EffecientB2_+_Bert.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "gpuClass": "standard",
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "5fb3ce547c094be5a95ab91dec35239b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_fac5192766354c6ea714135820ba9baa",
              "IPY_MODEL_1ee627fc31eb4038a96e8e6e74319fd8",
              "IPY_MODEL_f6c46660547e4b89aaee1771b8dfcd4e"
            ],
            "layout": "IPY_MODEL_2c6f50e07320456fadf95fcfffe82fe5"
          }
        },
        "fac5192766354c6ea714135820ba9baa": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_0ed4b7683c8f4148bdda6c9d8de0105f",
            "placeholder": "​",
            "style": "IPY_MODEL_6a97d46b971240fc89710ced32b4fa0f",
            "value": "Downloading (…)solve/main/vocab.txt: 100%"
          }
        },
        "1ee627fc31eb4038a96e8e6e74319fd8": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_19f403239d8c416992853077fb256169",
            "max": 231508,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_0bae9c1192004902ab3b6405ec0100c0",
            "value": 231508
          }
        },
        "f6c46660547e4b89aaee1771b8dfcd4e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_3b25874a1eb34e69b1a254ee3cd0dcaa",
            "placeholder": "​",
            "style": "IPY_MODEL_6d0260d6a39f479ea303252f5d4f5fac",
            "value": " 232k/232k [00:00&lt;00:00, 4.54MB/s]"
          }
        },
        "2c6f50e07320456fadf95fcfffe82fe5": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "0ed4b7683c8f4148bdda6c9d8de0105f": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "6a97d46b971240fc89710ced32b4fa0f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "19f403239d8c416992853077fb256169": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "0bae9c1192004902ab3b6405ec0100c0": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "3b25874a1eb34e69b1a254ee3cd0dcaa": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "6d0260d6a39f479ea303252f5d4f5fac": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "89dc8cdaff614483b8aa6bb9cffc1b37": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_94cd1383bbdf4be9beb5dc77313f88b2",
              "IPY_MODEL_ca69bb68b6144a60b3aa5345c564bced",
              "IPY_MODEL_737ac1589afd4dbbb38b5290b6e1a79f"
            ],
            "layout": "IPY_MODEL_4bd737b8c61e4f3087e03ed1a991b9d2"
          }
        },
        "94cd1383bbdf4be9beb5dc77313f88b2": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_9e7a3922a041437f9015174fe29fefcc",
            "placeholder": "​",
            "style": "IPY_MODEL_3882812a3169426886345f87b56e389c",
            "value": "Downloading (…)okenizer_config.json: 100%"
          }
        },
        "ca69bb68b6144a60b3aa5345c564bced": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_3c8ecd5fed1044aa9ef96c47898d0f25",
            "max": 28,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_13112cff5cbb4c44a0af17cab75009c1",
            "value": 28
          }
        },
        "737ac1589afd4dbbb38b5290b6e1a79f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_c211d7eda87e490db0588618def13950",
            "placeholder": "​",
            "style": "IPY_MODEL_d398c47e9a224f2b8d4027ba428feadf",
            "value": " 28.0/28.0 [00:00&lt;00:00, 1.31kB/s]"
          }
        },
        "4bd737b8c61e4f3087e03ed1a991b9d2": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "9e7a3922a041437f9015174fe29fefcc": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "3882812a3169426886345f87b56e389c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "3c8ecd5fed1044aa9ef96c47898d0f25": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "13112cff5cbb4c44a0af17cab75009c1": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "c211d7eda87e490db0588618def13950": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "d398c47e9a224f2b8d4027ba428feadf": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "1092e1708f4c4ac2bc69a8bff3112350": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_6e816e1c53684871bf0146478950995d",
              "IPY_MODEL_c24e60db68af4836b9832e838f6e7e29",
              "IPY_MODEL_95ff807dfc7e47c5a5afc7e94140d0ce"
            ],
            "layout": "IPY_MODEL_4c563d3c07dc4406a7bbcac048f659ac"
          }
        },
        "6e816e1c53684871bf0146478950995d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_ce687fe31c9543b4b981f20f89e4b172",
            "placeholder": "​",
            "style": "IPY_MODEL_85780fdfeeb04ffbb080a2f1c8437253",
            "value": "Downloading (…)lve/main/config.json: 100%"
          }
        },
        "c24e60db68af4836b9832e838f6e7e29": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_441f326476344d9da8475b50af4feeea",
            "max": 570,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_9e115f7eff0b4f6fb1fa0a1238c80c1f",
            "value": 570
          }
        },
        "95ff807dfc7e47c5a5afc7e94140d0ce": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_210bc13fa03e4abe8a3c6cf4b272fec6",
            "placeholder": "​",
            "style": "IPY_MODEL_70d7063610f64c96be56dc4446f164e3",
            "value": " 570/570 [00:00&lt;00:00, 33.0kB/s]"
          }
        },
        "4c563d3c07dc4406a7bbcac048f659ac": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "ce687fe31c9543b4b981f20f89e4b172": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "85780fdfeeb04ffbb080a2f1c8437253": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "441f326476344d9da8475b50af4feeea": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "9e115f7eff0b4f6fb1fa0a1238c80c1f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "210bc13fa03e4abe8a3c6cf4b272fec6": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "70d7063610f64c96be56dc4446f164e3": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "fd2c68dcf04746758952d4a83800db0b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_21b7113e6454424591ca81e4e6b0d849",
              "IPY_MODEL_a301a339beb5479cbde2c3659723e5a8",
              "IPY_MODEL_bde10fda39344fe2bd5fe57030acb2f7"
            ],
            "layout": "IPY_MODEL_7c855d452e554e83aa2d83a3fdb0cce6"
          }
        },
        "21b7113e6454424591ca81e4e6b0d849": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_bc1c306f2ca74d5aa7ab5dd894ef3a15",
            "placeholder": "​",
            "style": "IPY_MODEL_71ab4f0156e343c286fa125c24a9bfbc",
            "value": "Downloading tf_model.h5: 100%"
          }
        },
        "a301a339beb5479cbde2c3659723e5a8": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_58c9dbd342eb4183bfce38c8007a8f57",
            "max": 536063208,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_7569e61018144ff4a5287e6c3c5f24b3",
            "value": 536063208
          }
        },
        "bde10fda39344fe2bd5fe57030acb2f7": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_8e02377b4627418791bc440762a5dd90",
            "placeholder": "​",
            "style": "IPY_MODEL_0ad8638e33054feb88f120bf23be59e6",
            "value": " 536M/536M [00:33&lt;00:00, 15.9MB/s]"
          }
        },
        "7c855d452e554e83aa2d83a3fdb0cce6": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "bc1c306f2ca74d5aa7ab5dd894ef3a15": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "71ab4f0156e343c286fa125c24a9bfbc": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "58c9dbd342eb4183bfce38c8007a8f57": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "7569e61018144ff4a5287e6c3c5f24b3": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "8e02377b4627418791bc440762a5dd90": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "0ad8638e33054feb88f120bf23be59e6": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "49e1ccf9c1b4465688d7cb0d579b1407": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_4033b79a480545e097898a7187dc2ad7",
              "IPY_MODEL_3e44c7714327416a952b807129a0b3bb",
              "IPY_MODEL_eb0d4567906147dcac6ea7c8bfe0a004"
            ],
            "layout": "IPY_MODEL_99cf8da68dfe45e4ac0c905a51d56702"
          }
        },
        "4033b79a480545e097898a7187dc2ad7": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_2be8dce2e1b14c25ab6194ea8058d2bc",
            "placeholder": "​",
            "style": "IPY_MODEL_3193c4be62ab42c6808cafb01fb752cc",
            "value": "Downloading (…)rocessor_config.json: 100%"
          }
        },
        "3e44c7714327416a952b807129a0b3bb": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_550099681e7b47b584399dcb4f90d9ab",
            "max": 251,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_6d23bd90602945218d841e600f03db3e",
            "value": 251
          }
        },
        "eb0d4567906147dcac6ea7c8bfe0a004": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_369bf11acaa34be3a7ac89deb51c7b11",
            "placeholder": "​",
            "style": "IPY_MODEL_35c10214055045ce8805c82d412a5170",
            "value": " 251/251 [00:00&lt;00:00, 6.54kB/s]"
          }
        },
        "99cf8da68dfe45e4ac0c905a51d56702": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "2be8dce2e1b14c25ab6194ea8058d2bc": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "3193c4be62ab42c6808cafb01fb752cc": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "550099681e7b47b584399dcb4f90d9ab": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "6d23bd90602945218d841e600f03db3e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "369bf11acaa34be3a7ac89deb51c7b11": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "35c10214055045ce8805c82d412a5170": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "b49ca6508edb4bf28274e8095b1000b9": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_24853d948ce14530876b4c5a9f185cac",
              "IPY_MODEL_14b6c533183b4851ac0e3a3332f400bd",
              "IPY_MODEL_f42eee1160f3440487b67e443f636388"
            ],
            "layout": "IPY_MODEL_83af66f65cbf44a9860ead99d0c91c48"
          }
        },
        "24853d948ce14530876b4c5a9f185cac": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_95e285812f454193bca7bf2bef1dd9ad",
            "placeholder": "​",
            "style": "IPY_MODEL_b959a5e4809a4eca87e9a8d71ae1ddf2",
            "value": "Downloading (…)okenizer_config.json: 100%"
          }
        },
        "14b6c533183b4851ac0e3a3332f400bd": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_e0dd34a5ffed465c87241f556f493606",
            "max": 320,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_8f9c5a4231674632a45376ccce0e1140",
            "value": 320
          }
        },
        "f42eee1160f3440487b67e443f636388": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_2c5bd8dcef1b4d8aa9549d119c88b1a0",
            "placeholder": "​",
            "style": "IPY_MODEL_566d4bb4f15c402cb4e41969e48097dc",
            "value": " 320/320 [00:00&lt;00:00, 7.18kB/s]"
          }
        },
        "83af66f65cbf44a9860ead99d0c91c48": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "95e285812f454193bca7bf2bef1dd9ad": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "b959a5e4809a4eca87e9a8d71ae1ddf2": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "e0dd34a5ffed465c87241f556f493606": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "8f9c5a4231674632a45376ccce0e1140": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "2c5bd8dcef1b4d8aa9549d119c88b1a0": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "566d4bb4f15c402cb4e41969e48097dc": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "c0e41e70e5c94065906a7f58ab1438cc": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_e6bfef5cb040431997b8078391047ebf",
              "IPY_MODEL_ed682b7678cb49c6adbb675d7497eb7e",
              "IPY_MODEL_42d0f9bd693f4547b0a060786a392525"
            ],
            "layout": "IPY_MODEL_3911081e82f34606ac986fccbb424839"
          }
        },
        "e6bfef5cb040431997b8078391047ebf": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_8b33258fae294d40922819b776247572",
            "placeholder": "​",
            "style": "IPY_MODEL_f0e6503903014c19b3ac9a2103f106cf",
            "value": "Downloading (…)solve/main/vocab.txt: 100%"
          }
        },
        "ed682b7678cb49c6adbb675d7497eb7e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_cdffd49ceffe44ee9f430436274d23bb",
            "max": 231508,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_eb7ee9a281ec41f5bf8145fefb1fae27",
            "value": 231508
          }
        },
        "42d0f9bd693f4547b0a060786a392525": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_6f998e064cb949e4b6f36a058cd12b12",
            "placeholder": "​",
            "style": "IPY_MODEL_5e27afc0f30c43d7ba017df0d25d6ede",
            "value": " 232k/232k [00:00&lt;00:00, 2.57MB/s]"
          }
        },
        "3911081e82f34606ac986fccbb424839": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "8b33258fae294d40922819b776247572": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "f0e6503903014c19b3ac9a2103f106cf": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "cdffd49ceffe44ee9f430436274d23bb": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "eb7ee9a281ec41f5bf8145fefb1fae27": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "6f998e064cb949e4b6f36a058cd12b12": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "5e27afc0f30c43d7ba017df0d25d6ede": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "02a77e0692144cd295e8c1fe152062b9": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_29c113e742f4416992ba0e163979cc83",
              "IPY_MODEL_1a119ae9f75347edb31a9dea4c5a321c",
              "IPY_MODEL_4847264d0fcb4a3a8314be734f751a88"
            ],
            "layout": "IPY_MODEL_f609de01a1dd44519f87f6eca6085e7b"
          }
        },
        "29c113e742f4416992ba0e163979cc83": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_2f037d15541f421a91e485f8ed220cb0",
            "placeholder": "​",
            "style": "IPY_MODEL_ecb7ba0d62ea48f9aebc2f39683303e5",
            "value": "Downloading (…)/main/tokenizer.json: 100%"
          }
        },
        "1a119ae9f75347edb31a9dea4c5a321c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_a9a7c5db46e941e496d7819d9307edb7",
            "max": 466081,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_fc0ce5cd0bae47f8a2a5b37d1f106b45",
            "value": 466081
          }
        },
        "4847264d0fcb4a3a8314be734f751a88": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_cf03e0e5d23f4167a219263678dfc432",
            "placeholder": "​",
            "style": "IPY_MODEL_82d2d8cfedd648e89ea02a279bb5910c",
            "value": " 466k/466k [00:00&lt;00:00, 7.21MB/s]"
          }
        },
        "f609de01a1dd44519f87f6eca6085e7b": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "2f037d15541f421a91e485f8ed220cb0": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "ecb7ba0d62ea48f9aebc2f39683303e5": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "a9a7c5db46e941e496d7819d9307edb7": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "fc0ce5cd0bae47f8a2a5b37d1f106b45": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "cf03e0e5d23f4167a219263678dfc432": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "82d2d8cfedd648e89ea02a279bb5910c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "6dfe4178e81240bd9075c72f4403c036": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_239f5f9d3fcf4a89afe5c84073f95a08",
              "IPY_MODEL_1593ec48e067470eaafb7042c09f080e",
              "IPY_MODEL_bf966ae4f44b4ac0a5e372326cae885e"
            ],
            "layout": "IPY_MODEL_b6dfe9f0f307479bb19a25717b9fb58b"
          }
        },
        "239f5f9d3fcf4a89afe5c84073f95a08": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_dcb364b9559d4b8aa4866551e78b08ee",
            "placeholder": "​",
            "style": "IPY_MODEL_63db8f4a848c4de9a3bbf1d40f362ece",
            "value": "Downloading (…)cial_tokens_map.json: 100%"
          }
        },
        "1593ec48e067470eaafb7042c09f080e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_3aec61d1873f4611af0a21608f571469",
            "max": 112,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_7a146b8611bb46058633d84a3fbda022",
            "value": 112
          }
        },
        "bf966ae4f44b4ac0a5e372326cae885e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_9d075cb3c6234c0fa32dd06a43779c62",
            "placeholder": "​",
            "style": "IPY_MODEL_bc81c55cdce741cb967f7949afc1c0be",
            "value": " 112/112 [00:00&lt;00:00, 2.31kB/s]"
          }
        },
        "b6dfe9f0f307479bb19a25717b9fb58b": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "dcb364b9559d4b8aa4866551e78b08ee": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "63db8f4a848c4de9a3bbf1d40f362ece": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "3aec61d1873f4611af0a21608f571469": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "7a146b8611bb46058633d84a3fbda022": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "9d075cb3c6234c0fa32dd06a43779c62": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "bc81c55cdce741cb967f7949afc1c0be": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "5c551e7ab19345c78ef5fce4ad1a5d55": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_33bf2ec4030444da97e2170a7b495c42",
              "IPY_MODEL_04fa3ac3d16447e9ae327de9335b2916",
              "IPY_MODEL_0252733eae1d4a0ba341dbcc5a2c8385"
            ],
            "layout": "IPY_MODEL_c7f6c028471343d3be2c1ce7baae98b2"
          }
        },
        "33bf2ec4030444da97e2170a7b495c42": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_782a30615a9240a9abeacd55e72843ca",
            "placeholder": "​",
            "style": "IPY_MODEL_f40730cfa9d048138c56348127a97ee0",
            "value": "Downloading (…)lve/main/config.json: 100%"
          }
        },
        "04fa3ac3d16447e9ae327de9335b2916": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_21cb468281ad427895f348068c3b81a9",
            "max": 135543,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_1bf03756bde240f486501f817b459221",
            "value": 135543
          }
        },
        "0252733eae1d4a0ba341dbcc5a2c8385": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_1da6e6fbba124d9a9eb695f0b5a344c7",
            "placeholder": "​",
            "style": "IPY_MODEL_fbb31acccd2047f392287729fb93f051",
            "value": " 136k/136k [00:00&lt;00:00, 2.68MB/s]"
          }
        },
        "c7f6c028471343d3be2c1ce7baae98b2": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "782a30615a9240a9abeacd55e72843ca": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "f40730cfa9d048138c56348127a97ee0": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "21cb468281ad427895f348068c3b81a9": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "1bf03756bde240f486501f817b459221": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "1da6e6fbba124d9a9eb695f0b5a344c7": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "fbb31acccd2047f392287729fb93f051": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "e8ba25378e6b4ae0839a7c896b075358": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_7c1027bb1bd445539503a58cbb0f1058",
              "IPY_MODEL_31b8b82f3d5c4c89ab8ec6b282271770",
              "IPY_MODEL_3539ccad54ef43ba8618c7cbf7b151f4"
            ],
            "layout": "IPY_MODEL_1a7fc4ca59ab4de98f0b0505daf84255"
          }
        },
        "7c1027bb1bd445539503a58cbb0f1058": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_de0e88d785e5484798ffff60dd614233",
            "placeholder": "​",
            "style": "IPY_MODEL_fb07c5f562074c9185fc31b9d8678dab",
            "value": "Downloading pytorch_model.bin: 100%"
          }
        },
        "31b8b82f3d5c4c89ab8ec6b282271770": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_4cdc159e963c4a528f0a81aa0834aeb2",
            "max": 470435927,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_bd8301c50a9e46c6abd209c16d94795a",
            "value": 470435927
          }
        },
        "3539ccad54ef43ba8618c7cbf7b151f4": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_fb7aa1bc81a84e92b012fe5195bd8d13",
            "placeholder": "​",
            "style": "IPY_MODEL_dc849e5575564982b134e736ed0896fa",
            "value": " 470M/470M [00:29&lt;00:00, 16.3MB/s]"
          }
        },
        "1a7fc4ca59ab4de98f0b0505daf84255": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "de0e88d785e5484798ffff60dd614233": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "fb07c5f562074c9185fc31b9d8678dab": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "4cdc159e963c4a528f0a81aa0834aeb2": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "bd8301c50a9e46c6abd209c16d94795a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "fb7aa1bc81a84e92b012fe5195bd8d13": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "dc849e5575564982b134e736ed0896fa": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        }
      }
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "source": [
        "!pip install efficientnet"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "YhQgENtfGioV",
        "outputId": "85eaf6d9-1fa9-4d93-ed72-b1c6e5211cf7"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
            "Collecting efficientnet\n",
            "  Downloading efficientnet-1.1.1-py3-none-any.whl (18 kB)\n",
            "Collecting keras-applications<=1.0.8,>=1.0.7 (from efficientnet)\n",
            "  Downloading Keras_Applications-1.0.8-py3-none-any.whl (50 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m50.7/50.7 kB\u001b[0m \u001b[31m2.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: scikit-image in /usr/local/lib/python3.10/dist-packages (from efficientnet) (0.19.3)\n",
            "Requirement already satisfied: numpy>=1.9.1 in /usr/local/lib/python3.10/dist-packages (from keras-applications<=1.0.8,>=1.0.7->efficientnet) (1.22.4)\n",
            "Requirement already satisfied: h5py in /usr/local/lib/python3.10/dist-packages (from keras-applications<=1.0.8,>=1.0.7->efficientnet) (3.8.0)\n",
            "Requirement already satisfied: scipy>=1.4.1 in /usr/local/lib/python3.10/dist-packages (from scikit-image->efficientnet) (1.10.1)\n",
            "Requirement already satisfied: networkx>=2.2 in /usr/local/lib/python3.10/dist-packages (from scikit-image->efficientnet) (3.1)\n",
            "Requirement already satisfied: pillow!=7.1.0,!=7.1.1,!=8.3.0,>=6.1.0 in /usr/local/lib/python3.10/dist-packages (from scikit-image->efficientnet) (8.4.0)\n",
            "Requirement already satisfied: imageio>=2.4.1 in /usr/local/lib/python3.10/dist-packages (from scikit-image->efficientnet) (2.25.1)\n",
            "Requirement already satisfied: tifffile>=2019.7.26 in /usr/local/lib/python3.10/dist-packages (from scikit-image->efficientnet) (2023.4.12)\n",
            "Requirement already satisfied: PyWavelets>=1.1.1 in /usr/local/lib/python3.10/dist-packages (from scikit-image->efficientnet) (1.4.1)\n",
            "Requirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.10/dist-packages (from scikit-image->efficientnet) (23.1)\n",
            "Installing collected packages: keras-applications, efficientnet\n",
            "Successfully installed efficientnet-1.1.1 keras-applications-1.0.8\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install transformers"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yqMy6xJnKDQF",
        "outputId": "4763679d-3031-49a1-eaf2-466c26dc8a83"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
            "Collecting transformers\n",
            "  Downloading transformers-4.28.1-py3-none-any.whl (7.0 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m7.0/7.0 MB\u001b[0m \u001b[31m39.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: filelock in /usr/local/lib/python3.10/dist-packages (from transformers) (3.12.0)\n",
            "Collecting huggingface-hub<1.0,>=0.11.0 (from transformers)\n",
            "  Downloading huggingface_hub-0.14.1-py3-none-any.whl (224 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m224.5/224.5 kB\u001b[0m \u001b[31m11.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: numpy>=1.17 in /usr/local/lib/python3.10/dist-packages (from transformers) (1.22.4)\n",
            "Requirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.10/dist-packages (from transformers) (23.1)\n",
            "Requirement already satisfied: pyyaml>=5.1 in /usr/local/lib/python3.10/dist-packages (from transformers) (6.0)\n",
            "Requirement already satisfied: regex!=2019.12.17 in /usr/local/lib/python3.10/dist-packages (from transformers) (2022.10.31)\n",
            "Requirement already satisfied: requests in /usr/local/lib/python3.10/dist-packages (from transformers) (2.27.1)\n",
            "Collecting tokenizers!=0.11.3,<0.14,>=0.11.1 (from transformers)\n",
            "  Downloading tokenizers-0.13.3-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (7.8 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m7.8/7.8 MB\u001b[0m \u001b[31m56.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: tqdm>=4.27 in /usr/local/lib/python3.10/dist-packages (from transformers) (4.65.0)\n",
            "Requirement already satisfied: fsspec in /usr/local/lib/python3.10/dist-packages (from huggingface-hub<1.0,>=0.11.0->transformers) (2023.4.0)\n",
            "Requirement already satisfied: typing-extensions>=3.7.4.3 in /usr/local/lib/python3.10/dist-packages (from huggingface-hub<1.0,>=0.11.0->transformers) (4.5.0)\n",
            "Requirement already satisfied: urllib3<1.27,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests->transformers) (1.26.15)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests->transformers) (2022.12.7)\n",
            "Requirement already satisfied: charset-normalizer~=2.0.0 in /usr/local/lib/python3.10/dist-packages (from requests->transformers) (2.0.12)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests->transformers) (3.4)\n",
            "Installing collected packages: tokenizers, huggingface-hub, transformers\n",
            "Successfully installed huggingface-hub-0.14.1 tokenizers-0.13.3 transformers-4.28.1\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!wget https://s3.amazonaws.com/cvmlp/vqa/mscoco/vqa/v2_Questions_Train_mscoco.zip"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RcmseR3oInX8",
        "outputId": "5385412c-f13e-4322-9d43-764d7613cb59"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "--2023-05-09 22:01:59--  https://s3.amazonaws.com/cvmlp/vqa/mscoco/vqa/v2_Questions_Train_mscoco.zip\n",
            "Resolving s3.amazonaws.com (s3.amazonaws.com)... 52.217.95.88, 52.217.97.70, 52.216.216.96, ...\n",
            "Connecting to s3.amazonaws.com (s3.amazonaws.com)|52.217.95.88|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 7239401 (6.9M) [application/zip]\n",
            "Saving to: ‘v2_Questions_Train_mscoco.zip’\n",
            "\n",
            "v2_Questions_Train_ 100%[===================>]   6.90M  31.0MB/s    in 0.2s    \n",
            "\n",
            "2023-05-09 22:02:00 (31.0 MB/s) - ‘v2_Questions_Train_mscoco.zip’ saved [7239401/7239401]\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!unzip \"/content/v2_Questions_Train_mscoco.zip\""
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vsiMeJ_sInVM",
        "outputId": "3db20a80-f508-4d80-cf8d-e091498f0210"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Archive:  /content/v2_Questions_Train_mscoco.zip\n",
            "  inflating: v2_OpenEnded_mscoco_train2014_questions.json  \n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!wget https://s3.amazonaws.com/cvmlp/vqa/mscoco/vqa/v2_Annotations_Train_mscoco.zip"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Zv2ZY8MMInSs",
        "outputId": "cf4f7b09-c78e-451b-a1bc-50ce8427eb9e"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "--2023-05-09 22:02:00--  https://s3.amazonaws.com/cvmlp/vqa/mscoco/vqa/v2_Annotations_Train_mscoco.zip\n",
            "Resolving s3.amazonaws.com (s3.amazonaws.com)... 52.217.14.150, 52.217.230.40, 52.216.219.144, ...\n",
            "Connecting to s3.amazonaws.com (s3.amazonaws.com)|52.217.14.150|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 21708861 (21M) [application/zip]\n",
            "Saving to: ‘v2_Annotations_Train_mscoco.zip’\n",
            "\n",
            "v2_Annotations_Trai 100%[===================>]  20.70M  34.3MB/s    in 0.6s    \n",
            "\n",
            "2023-05-09 22:02:01 (34.3 MB/s) - ‘v2_Annotations_Train_mscoco.zip’ saved [21708861/21708861]\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!unzip \"/content/v2_Annotations_Train_mscoco.zip\""
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GYtTeTvuJBdG",
        "outputId": "b557fb32-bf96-45c7-ed65-8cba409fc4a3"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Archive:  /content/v2_Annotations_Train_mscoco.zip\n",
            "  inflating: v2_mscoco_train2014_annotations.json  \n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import json\n",
        "import tensorflow as tf\n",
        "from efficientnet.tfkeras import EfficientNetB2, preprocess_input\n",
        "from transformers import BertTokenizer, TFBertForQuestionAnswering\n",
        "import pickle\n",
        "import re\n",
        "import numpy as np\n",
        "import os\n",
        "from keras.preprocessing.image import ImageDataGenerator\n",
        "from keras.models import Model\n",
        "from keras.utils import load_img, img_to_array\n",
        "from tqdm import tqdm"
      ],
      "metadata": {
        "id": "QhOg4bpEMad7"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Read dictionary pkl file\n",
        "with open('/content/drive/MyDrive/FInal Project NTI/train_image_feature.pkl', 'rb') as fp:\n",
        "    train_imgs_features = pickle.load(fp)\n",
        "    print('successful')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xCtX9eKU00dF",
        "outputId": "7f992e99-1079-43ea-d16b-3e17c08accd7"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "successful\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_imgs_features"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "d9irg-g61cgK",
        "outputId": "949a38ab-02b8-4b18-bee9-83744ab64763"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{9: array([0.3641575 , 0.39742967, 0.848959  , ..., 0.07714461, 0.5495399 ,\n",
              "        0.5432545 ], dtype=float32),\n",
              " 25: array([0.2270625 , 0.3399307 , 0.25919005, ..., 0.18876874, 0.25891095,\n",
              "        0.1660124 ], dtype=float32),\n",
              " 30: array([0.28736764, 0.65010697, 1.5415224 , ..., 0.23868492, 1.1545415 ,\n",
              "        0.0908316 ], dtype=float32),\n",
              " 34: array([1.5141456 , 0.08357344, 0.03438409, ..., 0.54353356, 0.3973285 ,\n",
              "        0.02907126], dtype=float32),\n",
              " 36: array([1.0456916 , 0.17438439, 0.49074018, ..., 0.7129947 , 0.77394235,\n",
              "        0.05939685], dtype=float32),\n",
              " 49: array([0.19489115, 0.14706653, 0.16575038, ..., 0.19204912, 0.29227567,\n",
              "        0.5584221 ], dtype=float32),\n",
              " 61: array([0.03941727, 0.1623201 , 0.5579196 , ..., 0.77507114, 0.39080268,\n",
              "        0.05651416], dtype=float32),\n",
              " 64: array([0.07590349, 0.17269596, 0.14271513, ..., 0.9711013 , 0.1546471 ,\n",
              "        0.4162653 ], dtype=float32),\n",
              " 71: array([0.5848455 , 1.094417  , 0.06460974, ..., 0.08701868, 0.21991473,\n",
              "        0.01926157], dtype=float32),\n",
              " 72: array([0.16102093, 0.1187598 , 0.05496673, ..., 0.05736272, 0.4112905 ,\n",
              "        0.58711565], dtype=float32),\n",
              " 77: array([0.9306433 , 0.09834541, 0.44084534, ..., 1.0217161 , 0.40155357,\n",
              "        0.2785315 ], dtype=float32),\n",
              " 78: array([0.11313876, 0.1752441 , 0.7561891 , ..., 0.20451254, 0.16530728,\n",
              "        0.35656762], dtype=float32),\n",
              " 81: array([0.10561262, 0.1921048 , 0.03836066, ..., 0.2697366 , 0.12896842,\n",
              "        0.84102726], dtype=float32),\n",
              " 86: array([0.06234888, 0.10071263, 0.1398088 , ..., 0.5018903 , 0.47888878,\n",
              "        0.10326266], dtype=float32),\n",
              " 89: array([0.575547  , 0.66263247, 0.20272516, ..., 0.2349057 , 0.3217379 ,\n",
              "        0.7889204 ], dtype=float32),\n",
              " 92: array([0.11210473, 0.2655196 , 0.54627347, ..., 0.97620195, 0.18719852,\n",
              "        0.05059657], dtype=float32),\n",
              " 94: array([0.2949681 , 0.14402205, 0.02477363, ..., 0.4123438 , 0.49401748,\n",
              "        0.25467482], dtype=float32),\n",
              " 109: array([0.02372335, 0.27260262, 0.5220241 , ..., 0.2115951 , 0.2176134 ,\n",
              "        0.4925356 ], dtype=float32),\n",
              " 110: array([0.6738345 , 0.72697634, 0.04602013, ..., 0.41623998, 0.10180964,\n",
              "        0.23445025], dtype=float32),\n",
              " 113: array([0.3552475 , 0.557464  , 0.41520408, ..., 1.0864162 , 0.18808518,\n",
              "        0.9015113 ], dtype=float32),\n",
              " 127: array([0.08761559, 1.1500405 , 0.21014561, ..., 0.14339001, 0.10826682,\n",
              "        0.21255478], dtype=float32),\n",
              " 138: array([0.7993988 , 0.2185776 , 0.6396967 , ..., 0.14868036, 0.17698455,\n",
              "        0.9041765 ], dtype=float32),\n",
              " 142: array([0.20031126, 0.45696425, 0.5069052 , ..., 1.2209942 , 0.47424024,\n",
              "        0.        ], dtype=float32),\n",
              " 144: array([0.12036665, 0.38657105, 0.33983898, ..., 0.50337076, 0.18449491,\n",
              "        0.59177   ], dtype=float32),\n",
              " 149: array([0.8396679 , 0.00566414, 0.6692425 , ..., 0.21733092, 0.06442592,\n",
              "        0.2299821 ], dtype=float32),\n",
              " 151: array([0.293093  , 0.26633832, 0.38434154, ..., 0.1498595 , 0.29548857,\n",
              "        0.20105256], dtype=float32),\n",
              " 154: array([1.5093379 , 0.056858  , 0.18621245, ..., 0.83218133, 0.29277495,\n",
              "        0.04092358], dtype=float32),\n",
              " 165: array([0.4188358 , 1.2364241 , 1.0680375 , ..., 0.56748253, 0.19176279,\n",
              "        0.00826578], dtype=float32),\n",
              " 194: array([0.11022885, 0.65782404, 0.11444314, ..., 0.44751894, 0.37887383,\n",
              "        0.09154431], dtype=float32),\n",
              " 201: array([0.31538314, 0.89261454, 0.93447816, ..., 1.3571622 , 0.20702937,\n",
              "        0.30479255], dtype=float32),\n",
              " 247: array([0.01397731, 0.44499367, 0.06160954, ..., 0.15878713, 0.2639259 ,\n",
              "        0.20634091], dtype=float32),\n",
              " 250: array([0.253568  , 0.35954198, 0.07490596, ..., 0.6545367 , 0.5548632 ,\n",
              "        0.16656643], dtype=float32),\n",
              " 260: array([1.5419893 , 0.23746178, 0.14259884, ..., 0.11540451, 0.9824343 ,\n",
              "        0.08448927], dtype=float32),\n",
              " 263: array([0.1942204 , 0.17265472, 0.07988864, ..., 0.14811766, 0.13678187,\n",
              "        0.1775487 ], dtype=float32),\n",
              " 307: array([0.44045335, 0.50492954, 0.14766406, ..., 0.22389303, 0.21844536,\n",
              "        0.28869957], dtype=float32),\n",
              " 308: array([0.25293446, 0.48940945, 0.56699276, ..., 0.52170265, 1.280128  ,\n",
              "        0.55352163], dtype=float32),\n",
              " 309: array([0.274051  , 0.27311152, 0.3619886 , ..., 0.08315259, 0.04760282,\n",
              "        0.6630851 ], dtype=float32),\n",
              " 312: array([0.05761302, 0.3858005 , 0.05235665, ..., 0.39980492, 0.22152968,\n",
              "        0.34287405], dtype=float32),\n",
              " 315: array([0.43911842, 0.3867055 , 0.534261  , ..., 1.0336856 , 0.40904295,\n",
              "        0.01456753], dtype=float32),\n",
              " 321: array([0.08924957, 1.4374213 , 0.04467466, ..., 0.11561166, 0.32213068,\n",
              "        0.07195026], dtype=float32),\n",
              " 322: array([0.21199892, 0.0753141 , 0.17019647, ..., 0.07926413, 0.5392313 ,\n",
              "        0.07691398], dtype=float32),\n",
              " 326: array([0.1119354 , 0.12760378, 0.43859142, ..., 0.9630777 , 0.38221145,\n",
              "        0.3321447 ], dtype=float32),\n",
              " 332: array([0.06750146, 0.39768758, 0.08842906, ..., 0.16112605, 0.5157602 ,\n",
              "        0.04705362], dtype=float32),\n",
              " 349: array([0.845328  , 0.3899907 , 0.16465399, ..., 0.19096822, 0.384408  ,\n",
              "        0.00573748], dtype=float32),\n",
              " 368: array([0.16175508, 0.0393387 , 0.37299305, ..., 0.12710194, 0.15740986,\n",
              "        0.13689685], dtype=float32),\n",
              " 370: array([0.14710742, 0.23148242, 0.07439098, ..., 0.6629258 , 0.15048465,\n",
              "        0.17407392], dtype=float32),\n",
              " 382: array([0.04431813, 0.14004228, 0.31188488, ..., 1.1925292 , 0.35680977,\n",
              "        0.49007678], dtype=float32),\n",
              " 384: array([0.15947685, 0.21312681, 0.5914144 , ..., 0.40358073, 0.47608098,\n",
              "        0.00717144], dtype=float32),\n",
              " 389: array([2.0627089 , 0.39039338, 0.7249652 , ..., 0.38224465, 0.33042175,\n",
              "        0.31742197], dtype=float32),\n",
              " 394: array([0.13822708, 0.6152002 , 0.25154752, ..., 1.1239803 , 0.7292965 ,\n",
              "        0.52727914], dtype=float32),\n",
              " 404: array([0.23801452, 0.07262155, 0.3812902 , ..., 0.21331018, 0.79779065,\n",
              "        0.03981463], dtype=float32),\n",
              " 419: array([0.32153392, 0.8842347 , 0.21714075, ..., 0.10356477, 1.1329099 ,\n",
              "        0.7095115 ], dtype=float32),\n",
              " 431: array([0.18844673, 0.884076  , 0.3338266 , ..., 0.16673037, 0.00284826,\n",
              "        0.15013883], dtype=float32),\n",
              " 436: array([1.0394981 , 0.20363262, 0.45578352, ..., 0.46899766, 0.05481605,\n",
              "        0.20772937], dtype=float32),\n",
              " 438: array([0.22544104, 0.07144141, 0.17028867, ..., 0.17761719, 0.3557738 ,\n",
              "        0.32952705], dtype=float32),\n",
              " 443: array([0.5195527 , 0.17834249, 0.07380403, ..., 0.07159214, 0.34141698,\n",
              "        0.26203477], dtype=float32),\n",
              " 446: array([0.68594706, 0.10702267, 0.401066  , ..., 0.11620625, 0.8571322 ,\n",
              "        0.28080624], dtype=float32),\n",
              " 450: array([0.00927516, 1.1329995 , 0.07134274, ..., 0.8576836 , 0.665964  ,\n",
              "        0.19884753], dtype=float32),\n",
              " 471: array([0.12268327, 0.25296244, 0.24097654, ..., 0.52333325, 0.02508276,\n",
              "        0.5130061 ], dtype=float32),\n",
              " 490: array([0.38404045, 0.23725137, 0.34024787, ..., 0.5176542 , 0.5924278 ,\n",
              "        0.55069816], dtype=float32),\n",
              " 491: array([0.23473255, 0.2605238 , 0.8018326 , ..., 0.2838171 , 0.24238572,\n",
              "        0.29858422], dtype=float32),\n",
              " 508: array([0.14880455, 0.14485517, 0.28487372, ..., 0.3866472 , 0.02018755,\n",
              "        0.7507847 ], dtype=float32),\n",
              " 510: array([0.02935301, 0.06442493, 0.31043676, ..., 0.15311697, 0.09216525,\n",
              "        0.15674308], dtype=float32),\n",
              " 514: array([0.28742033, 0.2002894 , 0.79852355, ..., 0.12743016, 0.0127038 ,\n",
              "        0.5093146 ], dtype=float32),\n",
              " 529: array([0.20293802, 0.7957994 , 0.24505687, ..., 0.30414927, 0.39848652,\n",
              "        0.23546734], dtype=float32),\n",
              " 531: array([0.26051816, 0.1276479 , 0.15967149, ..., 0.3754994 , 0.18444999,\n",
              "        0.6636394 ], dtype=float32),\n",
              " 532: array([0.12750691, 0.08546007, 0.3243073 , ..., 0.5768734 , 0.10026545,\n",
              "        0.33413   ], dtype=float32),\n",
              " 540: array([0.16138898, 0.42376044, 0.03954659, ..., 0.07768999, 0.00498003,\n",
              "        0.24534239], dtype=float32),\n",
              " 542: array([0.14292979, 0.47785172, 0.16207895, ..., 0.30181193, 0.02374386,\n",
              "        0.5281931 ], dtype=float32),\n",
              " 560: array([0.3028291 , 0.1678819 , 0.3632145 , ..., 0.27521405, 0.90300107,\n",
              "        0.66866976], dtype=float32),\n",
              " 562: array([0.13140285, 0.12780736, 0.19129878, ..., 0.8553658 , 0.63045347,\n",
              "        0.7428919 ], dtype=float32),\n",
              " 572: array([0.415833  , 0.20883617, 0.34100085, ..., 0.02202775, 0.63093287,\n",
              "        0.38785344], dtype=float32),\n",
              " 575: array([0.5782863 , 0.6224443 , 0.01861669, ..., 0.37553412, 0.2275261 ,\n",
              "        0.08327593], dtype=float32),\n",
              " 581: array([0.48741573, 0.26131868, 0.16521169, ..., 0.25356925, 0.577494  ,\n",
              "        1.0163438 ], dtype=float32),\n",
              " 584: array([0.05728311, 0.5487008 , 0.15520325, ..., 0.05391476, 0.29023114,\n",
              "        0.67602843], dtype=float32),\n",
              " 595: array([0.0593667 , 0.03589087, 0.21292806, ..., 0.43111616, 0.11808159,\n",
              "        0.9035224 ], dtype=float32),\n",
              " 597: array([0.697803  , 0.27442968, 0.3894279 , ..., 0.1252583 , 0.04697035,\n",
              "        0.62968117], dtype=float32),\n",
              " 605: array([0.15929729, 0.5397066 , 0.4281724 , ..., 0.03562985, 0.29251033,\n",
              "        0.6222328 ], dtype=float32),\n",
              " 612: array([0.46536478, 0.54200953, 2.7542357 , ..., 0.29031858, 0.7910117 ,\n",
              "        1.6355649 ], dtype=float32),\n",
              " 620: array([0.19383979, 0.3417882 , 0.04058136, ..., 0.06629173, 0.21588099,\n",
              "        0.4961533 ], dtype=float32),\n",
              " 625: array([0.30864546, 0.3784841 , 0.33498734, ..., 0.38933346, 0.07747146,\n",
              "        0.4342532 ], dtype=float32),\n",
              " 629: array([0.37645382, 0.3857571 , 1.0833802 , ..., 0.38177377, 0.6814825 ,\n",
              "        0.3224302 ], dtype=float32),\n",
              " 634: array([0.08001842, 0.35848606, 0.40298754, ..., 0.5024618 , 0.7336546 ,\n",
              "        0.63012385], dtype=float32),\n",
              " 643: array([0.18914558, 0.07775985, 1.2797091 , ..., 0.64427954, 0.14874688,\n",
              "        0.2960388 ], dtype=float32),\n",
              " 650: array([0.36788142, 0.06679469, 0.55120456, ..., 0.7895056 , 1.4547913 ,\n",
              "        0.6175189 ], dtype=float32),\n",
              " 656: array([0.05037417, 0.28260803, 0.5600894 , ..., 0.07981721, 0.01056902,\n",
              "        0.19187304], dtype=float32),\n",
              " 659: array([0.12780632, 0.2068925 , 0.2040722 , ..., 0.07093222, 0.02127288,\n",
              "        0.2441994 ], dtype=float32),\n",
              " 670: array([0.1792647 , 0.4417089 , 0.07487088, ..., 0.08235066, 0.15311614,\n",
              "        0.07777666], dtype=float32),\n",
              " 671: array([0.17390075, 0.13964027, 0.09762494, ..., 0.13763413, 0.6035509 ,\n",
              "        0.8727361 ], dtype=float32),\n",
              " 673: array([0.17491779, 0.07979169, 0.3388739 , ..., 0.        , 0.73545265,\n",
              "        0.16053449], dtype=float32),\n",
              " 681: array([0.15533075, 0.15998307, 0.49536672, ..., 0.6320615 , 0.02747039,\n",
              "        0.39338797], dtype=float32),\n",
              " 684: array([0.40719026, 0.25155792, 1.0097029 , ..., 0.62668383, 1.0434955 ,\n",
              "        0.11566295], dtype=float32),\n",
              " 690: array([0.25952056, 0.16503105, 0.52753747, ..., 0.02701632, 0.02388163,\n",
              "        0.20565689], dtype=float32),\n",
              " 706: array([0.14032157, 0.13585429, 0.10095035, ..., 0.30691695, 0.57060647,\n",
              "        0.6887064 ], dtype=float32),\n",
              " 714: array([0.06754427, 0.10653859, 0.6137153 , ..., 0.02557795, 0.20395654,\n",
              "        0.49983513], dtype=float32),\n",
              " 716: array([0.18242964, 0.27170026, 0.26854005, ..., 0.0239497 , 0.7845632 ,\n",
              "        0.12293924], dtype=float32),\n",
              " 722: array([0.45827436, 0.11759594, 0.3248244 , ..., 0.18699285, 0.7294438 ,\n",
              "        0.12887123], dtype=float32),\n",
              " 723: array([1.1198328 , 0.19742292, 1.4621766 , ..., 0.17084658, 0.09333843,\n",
              "        0.5738673 ], dtype=float32),\n",
              " 731: array([0.2988358 , 0.14551224, 0.44597402, ..., 0.5483227 , 0.35810786,\n",
              "        0.13378154], dtype=float32),\n",
              " 735: array([0.14408633, 0.7431038 , 1.1189659 , ..., 0.56225693, 1.2627275 ,\n",
              "        0.8081177 ], dtype=float32),\n",
              " 753: array([0.07842373, 0.34466988, 0.08723199, ..., 0.09234847, 0.35487515,\n",
              "        0.16699207], dtype=float32),\n",
              " 754: array([0.16341618, 0.44479257, 0.1226145 , ..., 0.22738425, 1.309883  ,\n",
              "        0.2541229 ], dtype=float32),\n",
              " 762: array([0.45561188, 0.11713527, 0.30632788, ..., 0.03593194, 0.1161496 ,\n",
              "        0.30775464], dtype=float32),\n",
              " 781: array([0.43774173, 0.21669218, 0.41546533, ..., 0.22574848, 0.20232512,\n",
              "        0.02550199], dtype=float32),\n",
              " 790: array([1.3020968e+00, 5.8072986e-04, 8.8553798e-01, ..., 8.9521045e-03,\n",
              "        2.8698242e-01, 1.1861203e+00], dtype=float32),\n",
              " 795: array([1.7646142 , 0.28033364, 0.08131466, ..., 0.13014926, 0.10243639,\n",
              "        0.        ], dtype=float32),\n",
              " 797: array([0.04526148, 0.2963771 , 0.08157359, ..., 0.36599553, 0.12344344,\n",
              "        0.14294949], dtype=float32),\n",
              " 801: array([0.6688129 , 0.2754553 , 0.2625432 , ..., 0.1440795 , 0.0178531 ,\n",
              "        0.46321413], dtype=float32),\n",
              " 813: array([0.3801176 , 0.40896654, 0.04833686, ..., 0.04932076, 0.07842673,\n",
              "        0.75318944], dtype=float32),\n",
              " 821: array([0.367541  , 0.33714682, 0.563831  , ..., 0.32582533, 0.01543389,\n",
              "        0.63103056], dtype=float32),\n",
              " 825: array([0.16265485, 0.26934946, 0.40899655, ..., 0.5989994 , 0.16846263,\n",
              "        0.38785592], dtype=float32),\n",
              " 828: array([0.06946766, 0.03621669, 0.1353451 , ..., 0.13125876, 0.08017835,\n",
              "        0.36413252], dtype=float32),\n",
              " 839: array([0.18348595, 0.11667236, 0.17166227, ..., 0.03720267, 0.3060366 ,\n",
              "        0.31286922], dtype=float32),\n",
              " 853: array([0.13147977, 0.2573663 , 0.11340716, ..., 0.86154604, 0.24832219,\n",
              "        0.77150273], dtype=float32),\n",
              " 882: array([0.579964  , 1.24751   , 0.46074164, ..., 0.843532  , 0.65936244,\n",
              "        0.47868392], dtype=float32),\n",
              " 897: array([0.16911966, 0.15311393, 0.2635115 , ..., 0.4969897 , 0.84756726,\n",
              "        0.15736544], dtype=float32),\n",
              " 901: array([0.9864238 , 0.29398292, 0.1399738 , ..., 0.0214509 , 0.22654204,\n",
              "        0.10972717], dtype=float32),\n",
              " 902: array([0.01353659, 0.1480385 , 0.27950856, ..., 0.26772732, 0.18417238,\n",
              "        0.01555251], dtype=float32),\n",
              " 908: array([0.342964  , 0.22484772, 0.32214707, ..., 0.2279545 , 0.23226754,\n",
              "        0.99403393], dtype=float32),\n",
              " 909: array([0.05532616, 0.21630634, 0.84141755, ..., 0.37855282, 0.45614028,\n",
              "        0.04272253], dtype=float32),\n",
              " 913: array([0.12490804, 0.24714875, 0.24247335, ..., 0.060524  , 0.18883716,\n",
              "        0.0949551 ], dtype=float32),\n",
              " 925: array([0.45750305, 0.5751362 , 0.8689805 , ..., 0.09093142, 0.03649203,\n",
              "        0.03876643], dtype=float32),\n",
              " 927: array([0.36947155, 0.24282968, 1.4993737 , ..., 0.36102337, 0.43607718,\n",
              "        0.43756628], dtype=float32),\n",
              " 934: array([0.17389406, 0.21542287, 0.15556723, ..., 0.09163772, 0.7349206 ,\n",
              "        0.11077658], dtype=float32),\n",
              " 941: array([0.19487405, 0.06745844, 1.6827469 , ..., 0.6924029 , 0.5006268 ,\n",
              "        0.2541272 ], dtype=float32),\n",
              " 943: array([0.52232015, 0.47058412, 0.53591925, ..., 0.35587913, 0.5448899 ,\n",
              "        0.11197283], dtype=float32),\n",
              " 955: array([0.25103882, 0.32562453, 0.18184124, ..., 0.35708216, 0.332187  ,\n",
              "        0.4532293 ], dtype=float32),\n",
              " 960: array([1.351733  , 1.5950682 , 1.5592456 , ..., 0.08559892, 0.39522976,\n",
              "        0.28857213], dtype=float32),\n",
              " 965: array([0.3134837 , 0.2637664 , 0.3603336 , ..., 0.8292259 , 0.08418252,\n",
              "        1.5544728 ], dtype=float32),\n",
              " 977: array([0.30977988, 0.2918638 , 0.26793385, ..., 0.30085447, 0.14655301,\n",
              "        0.9451754 ], dtype=float32),\n",
              " 982: array([0.9254654 , 0.09858762, 0.17364602, ..., 0.33970064, 1.0335528 ,\n",
              "        0.21119592], dtype=float32),\n",
              " 984: array([0.09721234, 0.33394748, 1.0765481 , ..., 0.        , 0.05070037,\n",
              "        0.10615365], dtype=float32),\n",
              " 996: array([0.21276128, 0.16804628, 0.13403448, ..., 0.36642522, 0.07679047,\n",
              "        1.004425  ], dtype=float32),\n",
              " 1006: array([0.17058897, 0.34403628, 0.9004888 , ..., 0.33767545, 0.13034405,\n",
              "        0.12973276], dtype=float32),\n",
              " 1011: array([0.9578974 , 0.44909057, 0.6337526 , ..., 0.44417384, 0.0166869 ,\n",
              "        0.04138576], dtype=float32),\n",
              " 1014: array([0.6053378 , 0.16538915, 0.5421329 , ..., 0.48988613, 1.229406  ,\n",
              "        0.1883979 ], dtype=float32),\n",
              " 1025: array([0.09499775, 0.28107446, 0.19127095, ..., 0.05181083, 0.8515481 ,\n",
              "        0.24224104], dtype=float32),\n",
              " 1036: array([0.08053264, 0.54135764, 0.2702894 , ..., 0.32903248, 0.5294666 ,\n",
              "        0.06083978], dtype=float32),\n",
              " 1053: array([0.47495756, 0.41252762, 0.5391    , ..., 0.14536567, 0.3526618 ,\n",
              "        0.23309608], dtype=float32),\n",
              " 1059: array([0.01428353, 0.75506365, 0.        , ..., 1.1059585 , 0.43045226,\n",
              "        0.30467165], dtype=float32),\n",
              " 1072: array([0.08382457, 0.6067345 , 0.25974405, ..., 0.04982071, 0.18751565,\n",
              "        0.54604656], dtype=float32),\n",
              " 1084: array([0.69706297, 0.02483527, 0.18500786, ..., 0.22274005, 0.53019714,\n",
              "        0.18407655], dtype=float32),\n",
              " 1085: array([0.26629147, 0.2530727 , 0.2762691 , ..., 0.71969813, 0.05399255,\n",
              "        0.7696901 ], dtype=float32),\n",
              " 1090: array([0.4261615 , 0.45227757, 0.32521093, ..., 0.22849837, 0.23578788,\n",
              "        0.37706476], dtype=float32),\n",
              " 1098: array([0.7631097 , 1.0291681 , 0.26552054, ..., 0.4652413 , 0.940814  ,\n",
              "        1.0190952 ], dtype=float32),\n",
              " 1099: array([0.        , 0.01805701, 0.25423717, ..., 0.30431935, 0.7347213 ,\n",
              "        0.6097113 ], dtype=float32),\n",
              " 1102: array([0.1830499 , 0.09507102, 0.38656098, ..., 0.4677456 , 0.41318882,\n",
              "        0.33495116], dtype=float32),\n",
              " 1107: array([0.3097533 , 0.26573992, 0.25957763, ..., 0.98466647, 1.1157769 ,\n",
              "        0.34337285], dtype=float32),\n",
              " 1108: array([0.41368032, 0.24685577, 0.34174365, ..., 0.4849741 , 0.09171123,\n",
              "        0.        ], dtype=float32),\n",
              " 1111: array([0.06932506, 0.32693696, 0.03592486, ..., 0.14816707, 0.008005  ,\n",
              "        0.6345945 ], dtype=float32),\n",
              " 1122: array([0.47016478, 0.6639479 , 0.02745839, ..., 0.3515833 , 0.64734375,\n",
              "        0.0006693 ], dtype=float32),\n",
              " 1139: array([0.1807231 , 0.31280535, 0.35737637, ..., 0.26139432, 0.9113811 ,\n",
              "        0.6170721 ], dtype=float32),\n",
              " 1144: array([0.2710139 , 0.19818404, 1.303757  , ..., 0.5773982 , 0.25659657,\n",
              "        0.89034057], dtype=float32),\n",
              " 1145: array([0.7939355 , 0.05694517, 1.4468415 , ..., 0.6014272 , 0.26343787,\n",
              "        1.0569104 ], dtype=float32),\n",
              " 1155: array([0.14010994, 0.14280103, 0.34733236, ..., 0.08222193, 0.38743752,\n",
              "        0.14237371], dtype=float32),\n",
              " 1166: array([0.20000704, 0.06114295, 0.23991945, ..., 0.34520614, 0.66122055,\n",
              "        0.4892686 ], dtype=float32),\n",
              " 1168: array([0.03805885, 0.49254197, 0.01642467, ..., 0.22570266, 0.5120089 ,\n",
              "        0.4414177 ], dtype=float32),\n",
              " 1183: array([0.5945332 , 0.6215445 , 0.4798094 , ..., 0.3320843 , 0.5905113 ,\n",
              "        0.07294181], dtype=float32),\n",
              " 1200: array([0.8257015 , 0.03638929, 0.57505304, ..., 0.5487075 , 0.56419826,\n",
              "        0.14288336], dtype=float32),\n",
              " 1204: array([0.4195465 , 0.29887754, 0.9885059 , ..., 0.20143172, 0.35221654,\n",
              "        0.26358914], dtype=float32),\n",
              " 1213: array([0.04056865, 0.29140154, 0.03706034, ..., 0.00880921, 0.07038434,\n",
              "        0.7264533 ], dtype=float32),\n",
              " 1216: array([0.1222682 , 0.2083771 , 0.5712021 , ..., 1.0049944 , 0.29926607,\n",
              "        0.7197113 ], dtype=float32),\n",
              " 1224: array([0.6822755 , 0.2004578 , 0.19490547, ..., 1.7825058 , 0.67123884,\n",
              "        0.84165907], dtype=float32),\n",
              " 1232: array([0.12767312, 0.1423237 , 0.20922244, ..., 0.26158276, 0.00920819,\n",
              "        0.37436706], dtype=float32),\n",
              " 1237: array([0.797436  , 0.8086008 , 1.6678569 , ..., 1.099245  , 0.6105328 ,\n",
              "        0.29850733], dtype=float32),\n",
              " 1238: array([0.09600525, 0.28588563, 0.07745147, ..., 0.6929651 , 0.5003426 ,\n",
              "        0.209099  ], dtype=float32),\n",
              " 1261: array([0.10505927, 0.5697923 , 0.33634537, ..., 0.6402334 , 0.44409412,\n",
              "        0.28807968], dtype=float32),\n",
              " 1264: array([0.7819979 , 0.21615815, 0.6770199 , ..., 0.00651314, 0.26207185,\n",
              "        0.5281188 ], dtype=float32),\n",
              " 1271: array([0.65626514, 0.2083805 , 0.54026425, ..., 0.        , 0.31958038,\n",
              "        0.38575053], dtype=float32),\n",
              " 1282: array([0.20015568, 0.17820199, 0.3698119 , ..., 0.10537145, 0.14012678,\n",
              "        0.45347294], dtype=float32),\n",
              " 1295: array([0.53779376, 0.12006727, 0.80823815, ..., 0.7853231 , 0.25832528,\n",
              "        0.3302347 ], dtype=float32),\n",
              " 1298: array([0.11102851, 0.6656109 , 0.47069824, ..., 0.6598463 , 0.27526766,\n",
              "        1.3122356 ], dtype=float32),\n",
              " 1306: array([0.06099769, 0.09616071, 0.17461672, ..., 0.00173064, 0.34400576,\n",
              "        0.985999  ], dtype=float32),\n",
              " 1307: array([0.07374309, 0.4298311 , 0.88073766, ..., 0.6811937 , 0.6626715 ,\n",
              "        0.88546216], dtype=float32),\n",
              " 1308: array([0.31228673, 0.02123511, 0.09908026, ..., 0.00760549, 0.55833024,\n",
              "        0.6133095 ], dtype=float32),\n",
              " 1311: array([0.02355127, 0.25387183, 0.10484151, ..., 0.13859   , 0.5860277 ,\n",
              "        1.0833197 ], dtype=float32),\n",
              " 1315: array([0.47566783, 0.08707875, 0.5602236 , ..., 0.7146368 , 0.20594876,\n",
              "        0.00688809], dtype=float32),\n",
              " 1319: array([0.10919211, 0.04280772, 0.15042084, ..., 0.49531367, 0.9165374 ,\n",
              "        1.0080535 ], dtype=float32),\n",
              " 1323: array([0.28230447, 0.26337448, 0.17791057, ..., 0.7323177 , 0.61950433,\n",
              "        1.0704322 ], dtype=float32),\n",
              " 1330: array([0.1000485 , 0.18051173, 0.17960373, ..., 0.10867935, 0.6208608 ,\n",
              "        0.51078504], dtype=float32),\n",
              " 1332: array([0.02299565, 0.43554267, 0.23972446, ..., 0.4897794 , 0.5599512 ,\n",
              "        0.66478   ], dtype=float32),\n",
              " 1350: array([0.34046528, 0.32315466, 0.03899972, ..., 0.3535468 , 0.00929988,\n",
              "        0.1773924 ], dtype=float32),\n",
              " 1355: array([0.71706927, 0.09848473, 0.22980995, ..., 0.20455125, 0.52086985,\n",
              "        0.721601  ], dtype=float32),\n",
              " 1359: array([0.18901485, 0.24596941, 0.31627518, ..., 0.30281517, 0.10876443,\n",
              "        0.33118784], dtype=float32),\n",
              " 1360: array([0.42673337, 0.39685738, 0.7471875 , ..., 0.23542002, 1.1715178 ,\n",
              "        0.61524934], dtype=float32),\n",
              " 1366: array([0.18457158, 0.4174076 , 0.36370286, ..., 0.12364338, 0.10878603,\n",
              "        0.10795113], dtype=float32),\n",
              " 1375: array([0.16565566, 0.17310666, 0.34671766, ..., 1.0069335 , 0.77354443,\n",
              "        0.85903335], dtype=float32),\n",
              " 1381: array([0.39289922, 0.1122477 , 0.1650457 , ..., 0.20547384, 0.55711764,\n",
              "        0.29504663], dtype=float32),\n",
              " 1386: array([0.13786802, 0.7148056 , 0.9180927 , ..., 0.12568477, 0.7337281 ,\n",
              "        0.5864995 ], dtype=float32),\n",
              " 1390: array([0.17265499, 0.10272225, 0.11344393, ..., 0.33191288, 0.14639845,\n",
              "        0.24464296], dtype=float32),\n",
              " 1392: array([0.01049168, 0.09859058, 0.33219934, ..., 0.89114106, 1.2082738 ,\n",
              "        0.35546744], dtype=float32),\n",
              " 1397: array([0.29056978, 0.03442756, 0.18253314, ..., 0.1992786 , 0.83441955,\n",
              "        0.45020032], dtype=float32),\n",
              " 1401: array([0.5230247 , 0.10996655, 0.47046068, ..., 0.99844944, 0.17407379,\n",
              "        0.01624349], dtype=float32),\n",
              " 1403: array([0.10370457, 0.3992545 , 0.7853918 , ..., 0.8203634 , 0.77195764,\n",
              "        0.1323128 ], dtype=float32),\n",
              " 1407: array([0.25289276, 0.11006711, 0.3434231 , ..., 0.09441319, 0.1468272 ,\n",
              "        0.14441097], dtype=float32),\n",
              " 1408: array([0.17592989, 0.39169174, 0.24372333, ..., 0.2841007 , 0.29650807,\n",
              "        0.5458065 ], dtype=float32),\n",
              " 1424: array([0.37185156, 0.5136361 , 1.4466456 , ..., 0.0106018 , 0.20944688,\n",
              "        1.1348939 ], dtype=float32),\n",
              " 1431: array([0.10963966, 0.36786297, 0.24471709, ..., 0.22371435, 0.37449187,\n",
              "        0.6151573 ], dtype=float32),\n",
              " 1451: array([0.3035215 , 0.32760984, 0.37109265, ..., 0.3435315 , 0.45094782,\n",
              "        0.75500107], dtype=float32),\n",
              " 1453: array([0.17604002, 0.23246112, 0.44942006, ..., 0.01408311, 0.12522526,\n",
              "        1.266546  ], dtype=float32),\n",
              " 1455: array([0.11841373, 0.23144299, 0.35515326, ..., 0.33090788, 0.36864197,\n",
              "        2.0837677 ], dtype=float32),\n",
              " 1472: array([1.1860714 , 0.1972599 , 0.16277781, ..., 0.30026528, 0.7485988 ,\n",
              "        0.6684265 ], dtype=float32),\n",
              " 1488: array([0.1778378 , 0.17858014, 0.3059223 , ..., 0.40635926, 0.33555323,\n",
              "        0.07398591], dtype=float32),\n",
              " 1496: array([0.0666997 , 0.14302742, 0.10503712, ..., 0.13449854, 0.3199528 ,\n",
              "        0.2272129 ], dtype=float32),\n",
              " 1497: array([0.94459283, 0.0515353 , 0.5550025 , ..., 1.1716189 , 0.04154029,\n",
              "        0.5917648 ], dtype=float32),\n",
              " 1501: array([0.769274  , 0.45504692, 0.20048368, ..., 0.8381021 , 0.04713714,\n",
              "        0.44850135], dtype=float32),\n",
              " 1505: array([0.858889  , 0.23977785, 0.2459097 , ..., 0.64732826, 0.04905595,\n",
              "        0.7905322 ], dtype=float32),\n",
              " 1507: array([0.12257329, 0.4256042 , 0.3137455 , ..., 0.6245487 , 0.18145028,\n",
              "        0.51268244], dtype=float32),\n",
              " 1510: array([0.10774622, 0.48725033, 0.3332345 , ..., 0.35070524, 1.1536725 ,\n",
              "        0.08407009], dtype=float32),\n",
              " 1515: array([0.04734153, 0.32033122, 0.22902746, ..., 0.24203713, 0.        ,\n",
              "        0.04710669], dtype=float32),\n",
              " 1518: array([0.27296227, 0.12166988, 0.5769551 , ..., 1.0074053 , 0.17816585,\n",
              "        0.11283241], dtype=float32),\n",
              " 1522: array([0.44118977, 0.27060384, 0.1242963 , ..., 0.6247248 , 0.97912157,\n",
              "        0.05558161], dtype=float32),\n",
              " 1523: array([0.13520706, 0.23352484, 0.08157878, ..., 0.02187554, 0.08767071,\n",
              "        0.01038929], dtype=float32),\n",
              " 1526: array([0.02435091, 0.11870965, 0.63491344, ..., 0.61060953, 0.64669627,\n",
              "        0.6025959 ], dtype=float32),\n",
              " 1527: array([0.13694802, 0.35599685, 0.4642777 , ..., 0.33688828, 0.02281963,\n",
              "        0.495507  ], dtype=float32),\n",
              " 1536: array([0.14512977, 0.42333755, 0.01542369, ..., 0.37190136, 0.44266322,\n",
              "        0.07953787], dtype=float32),\n",
              " 1548: array([0.41392758, 0.05363338, 0.24915704, ..., 0.11590391, 0.03186303,\n",
              "        0.747228  ], dtype=float32),\n",
              " 1558: array([0.14182173, 0.6791025 , 1.3325459 , ..., 0.6418612 , 0.25104132,\n",
              "        0.6055912 ], dtype=float32),\n",
              " 1562: array([0.98773086, 0.18629922, 0.18217158, ..., 0.862554  , 0.3437665 ,\n",
              "        1.1831983 ], dtype=float32),\n",
              " 1569: array([0.2231905 , 0.35646534, 0.4622522 , ..., 0.19932184, 0.4087292 ,\n",
              "        0.1359789 ], dtype=float32),\n",
              " 1579: array([0.15401423, 0.45904702, 0.3582225 , ..., 0.22401598, 0.02033785,\n",
              "        0.01262802], dtype=float32),\n",
              " 1580: array([0.33682853, 0.28906476, 0.36608416, ..., 0.0542262 , 0.21960628,\n",
              "        0.28399235], dtype=float32),\n",
              " 1586: array([0.6190591 , 0.4505837 , 0.93779874, ..., 0.56168586, 0.41331255,\n",
              "        0.04646676], dtype=float32),\n",
              " 1589: array([0.4516263 , 0.94022876, 0.38023683, ..., 0.34174725, 1.2686169 ,\n",
              "        1.1771407 ], dtype=float32),\n",
              " 1596: array([0.36514357, 0.10574209, 0.04604687, ..., 0.37928528, 0.6026789 ,\n",
              "        0.06912158], dtype=float32),\n",
              " 1611: array([0.11521668, 0.31687105, 0.32548785, ..., 0.08141206, 0.00329925,\n",
              "        0.6755434 ], dtype=float32),\n",
              " 1622: array([0.05733358, 0.11267577, 0.06837524, ..., 0.6727758 , 0.6915162 ,\n",
              "        0.26759845], dtype=float32),\n",
              " 1625: array([0.16826406, 0.18985513, 0.14891875, ..., 0.78363544, 0.39815348,\n",
              "        0.38858145], dtype=float32),\n",
              " 1637: array([0.4962959 , 0.30963057, 0.80571425, ..., 0.87171674, 0.7837193 ,\n",
              "        0.28783238], dtype=float32),\n",
              " 1639: array([0.29682   , 0.18612604, 0.4137313 , ..., 0.00657474, 0.37704676,\n",
              "        0.5928255 ], dtype=float32),\n",
              " 1645: array([0.15160328, 0.20957644, 0.11506218, ..., 0.70180225, 0.29158044,\n",
              "        0.36083478], dtype=float32),\n",
              " 1670: array([0.18771018, 0.17235379, 0.40763238, ..., 0.29822856, 0.35993165,\n",
              "        0.73407924], dtype=float32),\n",
              " 1674: array([0.14773576, 1.4513757 , 0.48859462, ..., 0.65331256, 0.58938426,\n",
              "        0.17977118], dtype=float32),\n",
              " 1681: array([0.17694461, 0.1439552 , 0.33979023, ..., 0.25371823, 1.0858624 ,\n",
              "        0.19668455], dtype=float32),\n",
              " 1688: array([0.32136878, 0.2621733 , 0.17854044, ..., 0.42291802, 0.28906935,\n",
              "        0.14631039], dtype=float32),\n",
              " 1697: array([0.4028293 , 0.81165075, 0.525396  , ..., 1.8142076 , 0.37461644,\n",
              "        1.2379687 ], dtype=float32),\n",
              " 1706: array([0.303268  , 0.21298546, 0.14075464, ..., 0.05361691, 0.81200063,\n",
              "        0.75257075], dtype=float32),\n",
              " 1709: array([0.07360987, 0.5451504 , 0.44071594, ..., 0.60468024, 0.53676414,\n",
              "        0.27918652], dtype=float32),\n",
              " 1712: array([0.74211514, 0.34439892, 0.01473115, ..., 0.8449547 , 0.7958629 ,\n",
              "        0.00199402], dtype=float32),\n",
              " 1720: array([0.272037  , 0.06716067, 0.17560817, ..., 0.34745234, 0.4753484 ,\n",
              "        0.43795106], dtype=float32),\n",
              " 1732: array([0.22905351, 0.16768512, 0.2661678 , ..., 0.2261019 , 0.41647688,\n",
              "        0.14770487], dtype=float32),\n",
              " 1737: array([0.374699  , 0.12109884, 0.03841186, ..., 0.00524511, 0.54587036,\n",
              "        0.2500325 ], dtype=float32),\n",
              " 1756: array([0.39741755, 0.46255058, 0.8259896 , ..., 0.20064048, 1.2131028 ,\n",
              "        0.5649829 ], dtype=float32),\n",
              " 1762: array([0.29833728, 0.09172285, 0.20712024, ..., 0.07455502, 0.00703422,\n",
              "        0.10429087], dtype=float32),\n",
              " 1764: array([0.34914201, 0.4626151 , 0.22202815, ..., 0.22783622, 0.        ,\n",
              "        0.23425037], dtype=float32),\n",
              " 1771: array([0.07861646, 0.55309284, 0.06511194, ..., 0.80276525, 0.15081212,\n",
              "        0.29275158], dtype=float32),\n",
              " 1774: array([0.09042589, 0.150752  , 0.11247465, ..., 0.1286623 , 0.8284608 ,\n",
              "        0.40886194], dtype=float32),\n",
              " 1777: array([0.83745944, 0.09514087, 0.45408484, ..., 0.3628844 , 0.02257887,\n",
              "        0.10683867], dtype=float32),\n",
              " 1781: array([0.3560902 , 0.14531854, 1.7277458 , ..., 0.04572165, 0.28706884,\n",
              "        0.14815316], dtype=float32),\n",
              " 1785: array([0.32059845, 0.488306  , 0.18447167, ..., 0.87803745, 1.0242145 ,\n",
              "        0.21163183], dtype=float32),\n",
              " 1786: array([0.3987234 , 0.25042373, 0.53827083, ..., 0.20547995, 0.17597044,\n",
              "        0.18419276], dtype=float32),\n",
              " 1790: array([0.529669  , 0.3340191 , 0.40489584, ..., 0.03600748, 0.3202238 ,\n",
              "        0.40808037], dtype=float32),\n",
              " 1792: array([0.0624401 , 0.09572726, 0.19545238, ..., 0.23850189, 0.11544324,\n",
              "        0.19420514], dtype=float32),\n",
              " 1804: array([0.9223714 , 0.1482832 , 0.47981972, ..., 0.1130114 , 0.11952747,\n",
              "        0.4056098 ], dtype=float32),\n",
              " 1810: array([0.20359041, 0.26090434, 0.38217878, ..., 0.12827669, 0.12703884,\n",
              "        0.23699066], dtype=float32),\n",
              " 1811: array([0.06742586, 0.8341484 , 0.35597897, ..., 0.86733854, 1.0169995 ,\n",
              "        0.7174449 ], dtype=float32),\n",
              " 1813: array([0.09926546, 0.18349732, 0.14302795, ..., 1.1272438 , 0.20411919,\n",
              "        0.22126967], dtype=float32),\n",
              " 1815: array([0.14512143, 0.07198916, 0.05113301, ..., 0.04365532, 0.08914593,\n",
              "        0.54714286], dtype=float32),\n",
              " 1822: array([0.6795442 , 0.02580692, 0.41160345, ..., 0.1374875 , 0.21002333,\n",
              "        0.61693805], dtype=float32),\n",
              " 1837: array([0.67063   , 0.5277155 , 0.37423354, ..., 0.03842594, 0.83882487,\n",
              "        0.09016033], dtype=float32),\n",
              " 1864: array([0.50775546, 0.28539187, 0.50852716, ..., 0.04351319, 0.7869046 ,\n",
              "        0.13860331], dtype=float32),\n",
              " 1875: array([0.15924874, 0.20642674, 0.700125  , ..., 0.40026975, 0.15489125,\n",
              "        0.88948137], dtype=float32),\n",
              " 1877: array([0.14022687, 0.24545306, 0.3018442 , ..., 0.5825748 , 0.45831245,\n",
              "        0.18575397], dtype=float32),\n",
              " 1888: array([0.27209792, 0.31783146, 0.72147334, ..., 1.408525  , 0.94071305,\n",
              "        0.6409547 ], dtype=float32),\n",
              " 1895: array([0.01446322, 0.1805719 , 0.55879015, ..., 0.3224967 , 0.45233548,\n",
              "        0.08038582], dtype=float32),\n",
              " 1900: array([0.68183535, 0.00228071, 0.23105384, ..., 0.32999802, 0.7270394 ,\n",
              "        0.7112316 ], dtype=float32),\n",
              " 1902: array([0.01252363, 0.9383609 , 0.69163096, ..., 0.211279  , 0.14421014,\n",
              "        0.9001311 ], dtype=float32),\n",
              " 1906: array([0.38785   , 0.3325945 , 0.31057224, ..., 0.75138485, 0.12254098,\n",
              "        0.6406125 ], dtype=float32),\n",
              " 1907: array([1.1025178 , 0.07642364, 0.04803462, ..., 0.4682924 , 0.2848456 ,\n",
              "        0.02685039], dtype=float32),\n",
              " 1911: array([0.16559547, 0.27836522, 0.15426059, ..., 0.15016198, 0.65458965,\n",
              "        0.47209537], dtype=float32),\n",
              " 1912: array([2.2681417e-01, 2.6234347e-01, 3.3021018e-01, ..., 2.6808400e-04,\n",
              "        4.1616715e-02, 2.6575106e-01], dtype=float32),\n",
              " 1915: array([0.17785716, 0.23233661, 0.6774185 , ..., 0.04912635, 1.0013257 ,\n",
              "        0.14652556], dtype=float32),\n",
              " 1924: array([1.329293  , 0.15256307, 0.65752006, ..., 0.96718884, 0.22109471,\n",
              "        0.7534819 ], dtype=float32),\n",
              " 1926: array([0.2578921 , 0.36214483, 0.58842933, ..., 0.3238148 , 1.4343112 ,\n",
              "        0.44560578], dtype=float32),\n",
              " 1941: array([0.79755056, 0.3905662 , 0.39407444, ..., 1.2549729 , 0.38632768,\n",
              "        0.09643398], dtype=float32),\n",
              " 1942: array([0.27756643, 0.34980175, 0.46045232, ..., 0.30858013, 1.1456583 ,\n",
              "        0.37202877], dtype=float32),\n",
              " 1943: array([0.0951442 , 0.180236  , 0.21837032, ..., 0.07414545, 1.281764  ,\n",
              "        0.37942123], dtype=float32),\n",
              " 1947: array([0.09181224, 0.04418766, 0.59686756, ..., 0.19035265, 0.22814655,\n",
              "        0.23799294], dtype=float32),\n",
              " 1958: array([0.31482828, 0.51714206, 0.19352102, ..., 0.34630358, 0.21595229,\n",
              "        0.09554818], dtype=float32),\n",
              " 1966: array([0.30238783, 0.44460204, 0.24193028, ..., 0.21419057, 0.00226766,\n",
              "        0.29673934], dtype=float32),\n",
              " 1994: array([0.70707625, 0.31004423, 0.28661495, ..., 0.34580052, 0.6492386 ,\n",
              "        0.43694207], dtype=float32),\n",
              " 1997: array([0.35592815, 0.37314236, 0.32321772, ..., 0.37176085, 0.9796682 ,\n",
              "        0.01706193], dtype=float32),\n",
              " 1999: array([0.27492923, 0.32998562, 0.15476805, ..., 0.63071513, 0.7934599 ,\n",
              "        1.0513464 ], dtype=float32),\n",
              " 2001: array([0.47795042, 0.0441735 , 0.3187561 , ..., 0.14043084, 0.6539523 ,\n",
              "        0.00205289], dtype=float32),\n",
              " 2007: array([0.26415488, 0.13989317, 0.5712769 , ..., 0.23158643, 0.1595192 ,\n",
              "        0.09019193], dtype=float32),\n",
              " 2024: array([0.16422941, 0.2882213 , 0.2687167 , ..., 0.11647371, 0.9256633 ,\n",
              "        0.49427983], dtype=float32),\n",
              " 2055: array([0.1220779 , 0.28444383, 0.5874482 , ..., 0.1475709 , 0.09354767,\n",
              "        0.04223824], dtype=float32),\n",
              " 2056: array([0.27149144, 0.1637068 , 0.5081254 , ..., 0.03571805, 0.5202995 ,\n",
              "        0.20835367], dtype=float32),\n",
              " 2066: array([0.29897183, 0.36130768, 0.22721586, ..., 1.1383623 , 0.44482756,\n",
              "        0.88871074], dtype=float32),\n",
              " 2068: array([0.0229963 , 0.03885357, 0.3138132 , ..., 0.20972219, 0.39738503,\n",
              "        0.00067825], dtype=float32),\n",
              " 2072: array([0.16231824, 0.15967408, 0.6349315 , ..., 0.07231663, 0.1627087 ,\n",
              "        0.38334188], dtype=float32),\n",
              " 2083: array([0.4177203 , 0.33826128, 0.65391743, ..., 0.09058756, 0.5057697 ,\n",
              "        1.0108469 ], dtype=float32),\n",
              " 2089: array([0.66129345, 0.03851856, 0.41280746, ..., 0.17177197, 0.705093  ,\n",
              "        0.62534297], dtype=float32),\n",
              " 2093: array([0.37055773, 0.57371426, 0.2751075 , ..., 0.1141025 , 0.24031943,\n",
              "        0.842626  ], dtype=float32),\n",
              " 2106: array([0.08246772, 0.1546533 , 1.5339162 , ..., 0.8957704 , 0.87218565,\n",
              "        0.33303344], dtype=float32),\n",
              " 2114: array([0.5635114 , 0.03475481, 0.47129178, ..., 0.533512  , 0.4116055 ,\n",
              "        0.14805397], dtype=float32),\n",
              " 2135: array([0.49601895, 0.06825007, 0.8960289 , ..., 1.3120595 , 0.31057405,\n",
              "        0.10740405], dtype=float32),\n",
              " 2148: array([0.5056324 , 0.09137876, 0.13781953, ..., 0.8093252 , 0.2898992 ,\n",
              "        0.10716987], dtype=float32),\n",
              " 2150: array([0.22020039, 0.23688528, 1.5405357 , ..., 0.43933022, 0.35607547,\n",
              "        0.29537767], dtype=float32),\n",
              " 2178: array([0.10477555, 0.44865197, 0.00711913, ..., 0.23057549, 0.28142965,\n",
              "        1.2512984 ], dtype=float32),\n",
              " 2184: array([0.3653497 , 1.1333437 , 0.8859546 , ..., 0.52762246, 0.7261659 ,\n",
              "        0.2502414 ], dtype=float32),\n",
              " 2193: array([0.1829516 , 0.3213976 , 0.04313189, ..., 0.22448611, 0.04667237,\n",
              "        0.62659657], dtype=float32),\n",
              " 2197: array([0.03487536, 0.05951717, 0.37964404, ..., 0.21077143, 1.1040709 ,\n",
              "        1.6906083 ], dtype=float32),\n",
              " 2209: array([0.22230718, 0.01477821, 0.26408193, ..., 0.8370351 , 0.3699996 ,\n",
              "        0.17618221], dtype=float32),\n",
              " 2211: array([0.17154455, 0.03304219, 0.25455317, ..., 0.06705637, 0.6506727 ,\n",
              "        0.5603523 ], dtype=float32),\n",
              " 2217: array([0.11346926, 0.06668951, 0.39407098, ..., 0.00677733, 0.9834727 ,\n",
              "        0.761941  ], dtype=float32),\n",
              " 2229: array([0.6249784 , 0.04291793, 0.20849746, ..., 0.3513281 , 0.21890768,\n",
              "        0.08024853], dtype=float32),\n",
              " 2232: array([1.2611599 , 0.13793078, 0.04122567, ..., 0.47607255, 0.11394261,\n",
              "        0.07306545], dtype=float32),\n",
              " 2244: array([0.25644884, 0.03903541, 0.8329435 , ..., 0.1832884 , 0.40803528,\n",
              "        0.49456593], dtype=float32),\n",
              " 2258: array([0.30091578, 0.31308526, 0.1816159 , ..., 0.0281892 , 0.02544883,\n",
              "        0.16807342], dtype=float32),\n",
              " 2270: array([0.18966752, 0.19342935, 0.39286822, ..., 0.18320014, 0.29267952,\n",
              "        0.65174234], dtype=float32),\n",
              " 2276: array([0.898491  , 0.11706489, 0.43834242, ..., 0.81558233, 0.08622181,\n",
              "        0.8409096 ], dtype=float32),\n",
              " 2278: array([0.08097527, 0.50264466, 0.2567001 , ..., 0.1669269 , 0.44286838,\n",
              "        0.24605188], dtype=float32),\n",
              " 2279: array([0.06991841, 0.03131892, 0.27365053, ..., 0.25400123, 0.205776  ,\n",
              "        0.38586694], dtype=float32),\n",
              " 2280: array([0.10406499, 0.1922876 , 0.17064686, ..., 0.0431247 , 0.04423015,\n",
              "        0.6295745 ], dtype=float32),\n",
              " 2281: array([0.08060125, 0.15630688, 0.19316255, ..., 0.7367851 , 0.17739749,\n",
              "        0.34516093], dtype=float32),\n",
              " 2283: array([0.3909273 , 0.01476182, 0.19788244, ..., 0.15230355, 0.18581843,\n",
              "        0.66469806], dtype=float32),\n",
              " 2284: array([0.20156798, 0.16759323, 0.11644582, ..., 0.06926273, 0.24491063,\n",
              "        0.45095575], dtype=float32),\n",
              " 2296: array([0.92524457, 0.04199119, 0.59792143, ..., 0.        , 0.2913442 ,\n",
              "        0.700707  ], dtype=float32),\n",
              " 2309: array([0.07536276, 0.66748846, 0.88376963, ..., 0.47285408, 0.6021836 ,\n",
              "        0.2688176 ], dtype=float32),\n",
              " 2337: array([0.103353  , 0.76175255, 0.4057243 , ..., 0.64552486, 0.40054002,\n",
              "        0.26378328], dtype=float32),\n",
              " 2342: array([0.19968593, 0.4680521 , 0.23935255, ..., 0.18035747, 0.30076137,\n",
              "        0.04588952], dtype=float32),\n",
              " 2347: array([0.02315628, 0.07756862, 0.45648998, ..., 0.49955976, 0.32056156,\n",
              "        0.48276392], dtype=float32),\n",
              " 2349: array([0.34778017, 0.4275453 , 0.14443308, ..., 0.09875345, 0.24585733,\n",
              "        0.7123914 ], dtype=float32),\n",
              " 2369: array([0.24954033, 0.15573618, 0.2893619 , ..., 0.17404225, 0.06492426,\n",
              "        0.9833792 ], dtype=float32),\n",
              " 2372: array([0.23202628, 1.5105866 , 0.2235404 , ..., 0.8772092 , 0.18294075,\n",
              "        0.27297476], dtype=float32),\n",
              " 2374: array([0.02876531, 0.27592117, 0.02018547, ..., 1.8327281 , 1.2976301 ,\n",
              "        0.16526704], dtype=float32),\n",
              " 2377: array([0.7403501 , 0.46758512, 0.2370712 , ..., 0.45399392, 0.76837564,\n",
              "        0.6484828 ], dtype=float32),\n",
              " 2389: array([0.01274447, 0.1019568 , 0.76790714, ..., 0.71907717, 0.09171341,\n",
              "        0.03589298], dtype=float32),\n",
              " 2400: array([0.31534863, 0.7010368 , 0.06158265, ..., 0.3113158 , 1.3917674 ,\n",
              "        0.5094129 ], dtype=float32),\n",
              " 2402: array([0.3396458 , 0.12526348, 0.29430243, ..., 0.54263157, 0.43283516,\n",
              "        0.44619635], dtype=float32),\n",
              " 2411: array([0.20977539, 0.63681304, 0.47321886, ..., 0.39585048, 0.9163151 ,\n",
              "        0.38696146], dtype=float32),\n",
              " 2415: array([0.38870627, 0.38621503, 0.20506532, ..., 0.0854253 , 0.42491373,\n",
              "        0.83922637], dtype=float32),\n",
              " 2429: array([0.1954114 , 0.10027733, 0.10448281, ..., 0.02759542, 0.39708915,\n",
              "        0.8160668 ], dtype=float32),\n",
              " 2444: array([0.15774757, 0.5821233 , 0.2843306 , ..., 0.69677573, 0.29171437,\n",
              "        0.18261135], dtype=float32),\n",
              " 2445: array([0.15408273, 0.17100497, 0.05756487, ..., 0.00172251, 0.5876746 ,\n",
              "        0.07634315], dtype=float32),\n",
              " 2446: array([0.2042227 , 0.21991256, 0.11047316, ..., 0.05919908, 0.22834887,\n",
              "        0.03626579], dtype=float32),\n",
              " 2448: array([0.3410913 , 0.08452723, 0.6680851 , ..., 0.38110867, 0.03807588,\n",
              "        0.09240461], dtype=float32),\n",
              " 2451: array([0.23292832, 0.34670734, 0.09392681, ..., 0.11425461, 1.2895172 ,\n",
              "        0.20538504], dtype=float32),\n",
              " 2459: array([0.7372059 , 0.67889535, 0.3863106 , ..., 1.5018935 , 0.17811489,\n",
              "        1.1982597 ], dtype=float32),\n",
              " 2466: array([0.6223427 , 0.48594272, 1.3060389 , ..., 0.15472156, 0.10288425,\n",
              "        1.1269563 ], dtype=float32),\n",
              " 2470: array([0.10576577, 0.27688438, 0.2537131 , ..., 0.6140698 , 0.9911632 ,\n",
              "        0.61757016], dtype=float32),\n",
              " 2471: array([0.05601402, 0.20964259, 0.1373249 , ..., 0.37023103, 0.12134342,\n",
              "        0.08309943], dtype=float32),\n",
              " 2496: array([0.06567831, 0.41506717, 0.05223287, ..., 0.14983444, 0.22769782,\n",
              "        0.09995939], dtype=float32),\n",
              " 2498: array([0.15132529, 0.1788817 , 0.7852646 , ..., 0.07748221, 0.57397044,\n",
              "        0.567191  ], dtype=float32),\n",
              " 2531: array([0.71032953, 0.4742151 , 0.09120394, ..., 0.7008943 , 0.04074769,\n",
              "        0.49025756], dtype=float32),\n",
              " 2536: array([0.39441374, 0.0926936 , 0.0952486 , ..., 0.15486416, 0.6720507 ,\n",
              "        0.23830467], dtype=float32),\n",
              " 2543: array([0.1551019 , 0.22486152, 0.17386508, ..., 0.19292434, 0.08159794,\n",
              "        0.07847324], dtype=float32),\n",
              " 2544: array([0.31828034, 0.16721736, 0.24661615, ..., 0.15815285, 0.66176146,\n",
              "        0.3613743 ], dtype=float32),\n",
              " 2545: array([0.24624601, 0.45161438, 0.04806382, ..., 0.3300471 , 0.6051384 ,\n",
              "        0.02373869], dtype=float32),\n",
              " 2555: array([0.48790103, 0.08061443, 0.18857288, ..., 0.1786348 , 1.4223596 ,\n",
              "        0.01858957], dtype=float32),\n",
              " 2559: array([0.1127313 , 0.098121  , 0.16455685, ..., 0.07847269, 0.05003234,\n",
              "        0.23164165], dtype=float32),\n",
              " 2560: array([0.07853739, 0.24686511, 0.22809422, ..., 0.32780707, 0.05876487,\n",
              "        0.18175873], dtype=float32),\n",
              " 2563: array([0.46080595, 0.07794616, 0.05052342, ..., 0.12412749, 0.20042104,\n",
              "        0.08931579], dtype=float32),\n",
              " 2567: array([0.57114685, 0.45718592, 0.78430796, ..., 0.16479225, 0.7182545 ,\n",
              "        0.13349888], dtype=float32),\n",
              " 2570: array([0.22004089, 0.14675435, 0.16043413, ..., 0.2214342 , 0.01890651,\n",
              "        0.5488976 ], dtype=float32),\n",
              " 2575: array([0.47387818, 0.46649653, 0.4120524 , ..., 0.61887455, 0.24378812,\n",
              "        0.46015236], dtype=float32),\n",
              " 2583: array([0.11968274, 0.05677835, 0.18257016, ..., 0.2888223 , 0.8683619 ,\n",
              "        0.11787526], dtype=float32),\n",
              " 2585: array([0.71722436, 0.27876508, 0.46920583, ..., 0.4946251 , 0.1402575 ,\n",
              "        1.6175554 ], dtype=float32),\n",
              " 2591: array([1.5389385 , 0.38048804, 1.0330174 , ..., 0.7370899 , 0.23887506,\n",
              "        0.45388418], dtype=float32),\n",
              " 2602: array([0.1397605 , 0.6796427 , 0.19613189, ..., 0.04043493, 0.21075961,\n",
              "        0.11142202], dtype=float32),\n",
              " 2606: array([0.41166276, 0.16513205, 0.7249415 , ..., 0.7756125 , 0.26330167,\n",
              "        0.44191158], dtype=float32),\n",
              " 2608: array([0.06356308, 0.368782  , 0.25609812, ..., 0.06557592, 0.2369434 ,\n",
              "        0.13566902], dtype=float32),\n",
              " 2614: array([0.4365228 , 0.10606379, 1.2242477 , ..., 0.2511892 , 1.2361804 ,\n",
              "        0.07576121], dtype=float32),\n",
              " 2618: array([0.00346645, 0.08183508, 0.2545362 , ..., 0.33949777, 0.8891094 ,\n",
              "        0.7360558 ], dtype=float32),\n",
              " 2619: array([0.18056947, 0.05491418, 0.13226104, ..., 0.11629258, 0.04688136,\n",
              "        0.43854666], dtype=float32),\n",
              " 2623: array([0.12072632, 0.3586359 , 0.5938116 , ..., 0.02499406, 0.24927396,\n",
              "        0.7198677 ], dtype=float32),\n",
              " 2624: array([0.46831852, 0.570871  , 0.16718957, ..., 0.07618333, 0.01335311,\n",
              "        0.8886666 ], dtype=float32),\n",
              " 2639: array([0.47496668, 0.04379532, 0.06846185, ..., 0.09534529, 0.45791602,\n",
              "        0.04670786], dtype=float32),\n",
              " 2644: array([0.41405725, 0.14764178, 0.53120065, ..., 0.44496304, 0.22597367,\n",
              "        0.7600868 ], dtype=float32),\n",
              " 2645: array([0.3832265 , 0.4957697 , 0.78828454, ..., 0.56059146, 0.08229782,\n",
              "        0.21460563], dtype=float32),\n",
              " 2658: array([0.36573482, 0.18135694, 0.38328558, ..., 0.13618563, 0.35227358,\n",
              "        1.2717652 ], dtype=float32),\n",
              " 2664: array([0.31503925, 0.16141972, 0.03779043, ..., 0.02501605, 0.04513424,\n",
              "        0.12589778], dtype=float32),\n",
              " 2672: array([0.68265903, 0.71558726, 0.23320273, ..., 1.0053589 , 0.5988822 ,\n",
              "        0.00224257], dtype=float32),\n",
              " 2686: array([0.22538489, 0.16162384, 0.3345295 , ..., 0.6366533 , 0.0686026 ,\n",
              "        0.594115  ], dtype=float32),\n",
              " 2687: array([0.0845468 , 0.04479448, 0.40549234, ..., 1.0897288 , 0.3732825 ,\n",
              "        0.1011811 ], dtype=float32),\n",
              " 2691: array([0.2566309 , 0.18068115, 0.5443359 , ..., 0.14935908, 0.5866955 ,\n",
              "        0.5881228 ], dtype=float32),\n",
              " 2693: array([0.7121968 , 0.2550059 , 0.30504328, ..., 0.18018141, 0.47479704,\n",
              "        0.28279936], dtype=float32),\n",
              " 2697: array([0.13487169, 0.7006257 , 0.7702793 , ..., 0.09759657, 0.26689446,\n",
              "        0.11132702], dtype=float32),\n",
              " 2703: array([0.19552243, 0.9762608 , 0.25018626, ..., 0.5287135 , 1.3739114 ,\n",
              "        0.6428478 ], dtype=float32),\n",
              " 2732: array([0.20438023, 0.5019372 , 0.05997186, ..., 0.343917  , 0.5984259 ,\n",
              "        0.00239252], dtype=float32),\n",
              " 2742: array([0.41232318, 0.6698694 , 0.3025216 , ..., 0.5113718 , 0.00733176,\n",
              "        0.34564018], dtype=float32),\n",
              " 2752: array([1.0244999 , 0.01846256, 0.74402535, ..., 0.20654556, 0.08149743,\n",
              "        0.6124958 ], dtype=float32),\n",
              " 2754: array([0.5829074 , 0.11583464, 0.49572268, ..., 0.34075162, 0.07895229,\n",
              "        0.30448252], dtype=float32),\n",
              " 2755: array([0.11066335, 0.17201835, 0.09066892, ..., 0.01563191, 0.4312544 ,\n",
              "        0.05195783], dtype=float32),\n",
              " 2758: array([0.01066269, 0.14350596, 0.06429531, ..., 0.51882994, 0.76927626,\n",
              "        0.35744393], dtype=float32),\n",
              " 2770: array([0.30142134, 0.32513586, 0.00499783, ..., 0.11344166, 0.05749951,\n",
              "        0.11898834], dtype=float32),\n",
              " 2774: array([0.68230534, 0.28095886, 0.78133756, ..., 0.0535133 , 0.21032137,\n",
              "        0.54565513], dtype=float32),\n",
              " 2776: array([0.39881608, 0.1405086 , 0.32561332, ..., 0.5405699 , 0.56190985,\n",
              "        0.40025115], dtype=float32),\n",
              " 2782: array([0.4531485 , 0.22891621, 0.76739156, ..., 0.1851463 , 0.78235877,\n",
              "        0.00176738], dtype=float32),\n",
              " 2823: array([0.02604649, 0.31752384, 0.3374354 , ..., 0.26005495, 0.87525964,\n",
              "        0.6047497 ], dtype=float32),\n",
              " 2833: array([0.29129234, 0.3330878 , 0.6542881 , ..., 0.05223636, 0.473953  ,\n",
              "        0.07517813], dtype=float32),\n",
              " 2842: array([0.27497435, 0.31940633, 0.2638054 , ..., 0.53062356, 0.35242367,\n",
              "        0.16192536], dtype=float32),\n",
              " 2843: array([0.09809905, 0.10008328, 0.43332106, ..., 1.1091795 , 0.7485374 ,\n",
              "        0.18163474], dtype=float32),\n",
              " 2849: array([0.10950024, 0.21833453, 0.21870472, ..., 0.2301096 , 0.2637595 ,\n",
              "        0.05671664], dtype=float32),\n",
              " 2860: array([0.27329373, 0.23860113, 0.29819053, ..., 0.7745427 , 0.8431544 ,\n",
              "        0.4721142 ], dtype=float32),\n",
              " 2886: array([0.13281351, 0.6542947 , 1.3870077 , ..., 0.20042053, 0.58738774,\n",
              "        0.4413116 ], dtype=float32),\n",
              " 2892: array([0.5668306 , 0.2826628 , 0.0725162 , ..., 0.52031463, 0.05950323,\n",
              "        0.34474707], dtype=float32),\n",
              " 2896: array([0.20804852, 0.3264619 , 0.2104534 , ..., 1.3148148 , 0.57523876,\n",
              "        0.13028905], dtype=float32),\n",
              " 2902: array([0.1341154 , 0.0822535 , 0.12448355, ..., 0.13747212, 0.21024132,\n",
              "        0.16301586], dtype=float32),\n",
              " 2907: array([0.22157617, 0.0763309 , 0.2650508 , ..., 0.02406868, 0.33157957,\n",
              "        0.40788725], dtype=float32),\n",
              " 2931: array([0.09837382, 0.1442078 , 0.5034099 , ..., 0.51926917, 0.12371221,\n",
              "        0.12196527], dtype=float32),\n",
              " 2951: array([0.7155268 , 0.08120024, 0.893286  , ..., 0.20626049, 0.49512184,\n",
              "        0.2252152 ], dtype=float32),\n",
              " 2963: array([0.2698779 , 0.36853787, 0.36108068, ..., 1.7359686 , 1.2243137 ,\n",
              "        0.11223003], dtype=float32),\n",
              " 2964: array([0.03334625, 0.23748864, 0.15176058, ..., 0.1436405 , 0.20298404,\n",
              "        0.30978674], dtype=float32),\n",
              " 2982: array([0.52935195, 0.8518236 , 0.05140208, ..., 0.04527379, 0.20256792,\n",
              "        0.04293653], dtype=float32),\n",
              " 2983: array([0.41077447, 0.7020173 , 0.62679964, ..., 0.27177337, 0.4457546 ,\n",
              "        0.01115004], dtype=float32),\n",
              " 2989: array([0.38517815, 0.5245015 , 0.8080951 , ..., 1.3506057 , 0.13325886,\n",
              "        0.19989382], dtype=float32),\n",
              " 2992: array([0.1901527 , 0.508782  , 0.21589163, ..., 0.2635365 , 0.06845725,\n",
              "        0.42757505], dtype=float32),\n",
              " 2998: array([0.32585686, 0.2961083 , 0.36263523, ..., 0.6384344 , 0.55009615,\n",
              "        0.14807156], dtype=float32),\n",
              " 3000: array([0.34402645, 0.72632784, 1.0414321 , ..., 1.0548317 , 0.32451588,\n",
              "        0.55879146], dtype=float32),\n",
              " 3003: array([0.20124766, 0.52316517, 0.13846597, ..., 0.63888574, 0.48592716,\n",
              "        0.00324607], dtype=float32),\n",
              " 3008: array([1.1513264 , 0.8670515 , 1.1893069 , ..., 0.23193696, 1.2853045 ,\n",
              "        0.43897408], dtype=float32),\n",
              " 3040: array([0.278124  , 0.40343374, 0.53981125, ..., 0.56820124, 0.7618063 ,\n",
              "        0.00451324], dtype=float32),\n",
              " 3048: array([0.13627085, 0.22757383, 0.26393628, ..., 0.2311314 , 0.1654312 ,\n",
              "        0.26653185], dtype=float32),\n",
              " 3076: array([0.1420729 , 0.18779984, 0.14630711, ..., 0.43717986, 0.6110015 ,\n",
              "        0.09972285], dtype=float32),\n",
              " 3077: array([0.38269258, 0.07182336, 0.38125724, ..., 0.01486375, 0.055483  ,\n",
              "        0.93992555], dtype=float32),\n",
              " 3080: array([1.1387932 , 0.39839968, 0.38720575, ..., 0.06795993, 0.54376435,\n",
              "        0.20251764], dtype=float32),\n",
              " 3118: array([0.22410753, 0.29265654, 0.4695852 , ..., 0.11220649, 0.19209397,\n",
              "        0.05968774], dtype=float32),\n",
              " 3124: array([0.16533567, 0.24441007, 0.48841584, ..., 0.07956128, 0.31158164,\n",
              "        0.7282889 ], dtype=float32),\n",
              " 3131: array([0.24669483, 0.81214416, 0.06241721, ..., 0.13233763, 0.71331483,\n",
              "        0.16146939], dtype=float32),\n",
              " 3148: array([0.43347648, 0.07390226, 0.13457555, ..., 0.07874453, 0.25143835,\n",
              "        0.04190657], dtype=float32),\n",
              " 3157: array([0.14004366, 0.067872  , 0.11249882, ..., 0.0178571 , 0.27389407,\n",
              "        0.30662596], dtype=float32),\n",
              " 3160: array([0.35700074, 0.41287583, 0.95090985, ..., 0.18427426, 0.07808296,\n",
              "        0.8796637 ], dtype=float32),\n",
              " 3178: array([0.37272507, 0.8736864 , 0.18890177, ..., 0.39598644, 0.27197164,\n",
              "        0.81426513], dtype=float32),\n",
              " 3197: array([0.39425576, 0.03595513, 0.74295205, ..., 0.49167302, 1.7796597 ,\n",
              "        0.6019739 ], dtype=float32),\n",
              " 3219: array([0.74889183, 0.02175057, 0.1463513 , ..., 0.35695183, 0.22408631,\n",
              "        0.6852956 ], dtype=float32),\n",
              " 3220: array([0.22810584, 0.13955246, 0.17361327, ..., 0.23415428, 0.00049427,\n",
              "        0.0440334 ], dtype=float32),\n",
              " 3224: array([0.28282735, 0.10285133, 0.30306104, ..., 0.08717006, 0.15096965,\n",
              "        0.08734111], dtype=float32),\n",
              " 3225: array([0.12581384, 0.633577  , 0.41580456, ..., 0.49382594, 0.21326184,\n",
              "        0.26443517], dtype=float32),\n",
              " 3234: array([0.11166722, 0.34849858, 0.295628  , ..., 0.4686632 , 0.88675797,\n",
              "        0.28355634], dtype=float32),\n",
              " 3236: array([0.5651672 , 0.03068178, 0.7401769 , ..., 0.        , 0.5010383 ,\n",
              "        0.29798973], dtype=float32),\n",
              " 3242: array([0.12668973, 0.17858842, 0.532388  , ..., 0.32484478, 0.51364964,\n",
              "        0.07170336], dtype=float32),\n",
              " 3249: array([0.3004418 , 0.16927381, 0.16133656, ..., 0.06128966, 0.56828976,\n",
              "        0.8669528 ], dtype=float32),\n",
              " 3259: array([0.01160806, 0.06472464, 0.8706498 , ..., 0.4623784 , 0.64569104,\n",
              "        0.14084187], dtype=float32),\n",
              " 3264: array([0.18626551, 0.20103748, 0.19723526, ..., 0.05523244, 0.8886572 ,\n",
              "        0.01305531], dtype=float32),\n",
              " 3270: array([0.3041595 , 0.39548987, 0.49287003, ..., 0.05196488, 0.18221773,\n",
              "        0.27943826], dtype=float32),\n",
              " 3272: array([0.5570741 , 0.227873  , 0.5249185 , ..., 0.13105631, 0.08347863,\n",
              "        0.09348428], dtype=float32),\n",
              " 3276: array([0.6330304 , 0.22310278, 0.6436039 , ..., 0.72341347, 1.0234842 ,\n",
              "        0.22504039], dtype=float32),\n",
              " 3286: array([0.69240785, 0.12789805, 0.4737405 , ..., 0.35633683, 0.92301387,\n",
              "        0.34451133], dtype=float32),\n",
              " 3293: array([0.1531361 , 0.51006794, 0.45934343, ..., 0.0893259 , 0.05383287,\n",
              "        1.0386444 ], dtype=float32),\n",
              " 3305: array([0.07962058, 0.51886725, 0.1358873 , ..., 0.19145097, 0.5612139 ,\n",
              "        0.23552826], dtype=float32),\n",
              " 3314: array([0.48856065, 0.10054184, 0.50964236, ..., 0.07564093, 0.03335226,\n",
              "        0.26642016], dtype=float32),\n",
              " 3320: array([0.13856077, 0.27597943, 0.17250934, ..., 1.4525347 , 0.90633196,\n",
              "        0.49308723], dtype=float32),\n",
              " 3321: array([0.20742299, 0.32303154, 0.21107644, ..., 0.50244653, 0.11050797,\n",
              "        0.5634783 ], dtype=float32),\n",
              " 3325: array([0.19201982, 0.09576154, 0.08957045, ..., 0.7202232 , 0.4256022 ,\n",
              "        0.12893596], dtype=float32),\n",
              " 3348: array([0.657141  , 0.7101048 , 0.7135087 , ..., 0.9986    , 0.16895935,\n",
              "        0.02957305], dtype=float32),\n",
              " 3353: array([0.20461762, 0.16190377, 0.51059693, ..., 0.01623431, 0.085277  ,\n",
              "        0.5842166 ], dtype=float32),\n",
              " 3361: array([0.0630891 , 0.06119049, 0.09568992, ..., 0.77228606, 0.04895853,\n",
              "        0.21604592], dtype=float32),\n",
              " 3365: array([0.34852907, 0.20786941, 0.475661  , ..., 0.16467634, 0.54377145,\n",
              "        0.18807673], dtype=float32),\n",
              " 3366: array([0.16961795, 0.22263406, 0.28407928, ..., 0.69235826, 0.2402607 ,\n",
              "        0.42284334], dtype=float32),\n",
              " 3375: array([0.22777875, 0.3985391 , 0.19276997, ..., 0.22613758, 0.21344024,\n",
              "        0.4887398 ], dtype=float32),\n",
              " 3386: array([0.42138067, 0.42527193, 0.2428295 , ..., 0.37916118, 0.21331185,\n",
              "        0.21556875], dtype=float32),\n",
              " 3389: array([0.13261873, 0.5868077 , 0.12032568, ..., 0.29501233, 0.08170445,\n",
              "        0.03148232], dtype=float32),\n",
              " 3398: array([0.25452408, 0.13044843, 0.03064655, ..., 0.0260795 , 0.01020327,\n",
              "        0.22466233], dtype=float32),\n",
              " 3412: array([0.13329422, 0.23079242, 0.6867478 , ..., 0.46937248, 1.0336833 ,\n",
              "        1.0601906 ], dtype=float32),\n",
              " 3432: array([0.20968981, 0.34338292, 0.3029635 , ..., 0.37251833, 0.5988549 ,\n",
              "        0.6748477 ], dtype=float32),\n",
              " 3442: array([0.23427045, 0.5229685 , 0.07208678, ..., 0.6794564 , 0.01986215,\n",
              "        0.16305923], dtype=float32),\n",
              " 3457: array([0.06523724, 0.24809352, 0.15698498, ..., 0.25175366, 0.75262445,\n",
              "        0.24345466], dtype=float32),\n",
              " 3461: array([0.21759848, 0.08477922, 0.3974768 , ..., 0.35836667, 0.37398604,\n",
              "        0.3880779 ], dtype=float32),\n",
              " 3464: array([0.11940918, 0.20848787, 1.9607594 , ..., 0.5161158 , 0.4905301 ,\n",
              "        0.17295209], dtype=float32),\n",
              " 3474: array([0.71175456, 0.30959523, 0.28756282, ..., 0.49606624, 0.7453582 ,\n",
              "        0.14512622], dtype=float32),\n",
              " 3478: array([0.10721646, 0.28658572, 0.49196023, ..., 0.43082714, 0.37497205,\n",
              "        0.09996202], dtype=float32),\n",
              " 3481: array([0.36029106, 0.431679  , 0.12394859, ..., 0.26006502, 0.10131434,\n",
              "        0.02622166], dtype=float32),\n",
              " 3483: array([0.81405044, 0.08070879, 0.6427478 , ..., 0.59006244, 0.01721165,\n",
              "        0.6675386 ], dtype=float32),\n",
              " 3493: array([0.59169906, 0.03097757, 0.53231233, ..., 0.95650226, 0.25436398,\n",
              "        0.45185655], dtype=float32),\n",
              " 3511: array([0.09828548, 0.41348562, 0.29565307, ..., 0.39175838, 1.3313735 ,\n",
              "        0.43171054], dtype=float32),\n",
              " 3514: array([0.3539585 , 0.9357159 , 0.28355855, ..., 0.1624322 , 0.711479  ,\n",
              "        0.3825802 ], dtype=float32),\n",
              " 3517: array([0.16426773, 0.30702943, 0.07168809, ..., 0.4185258 , 0.05458807,\n",
              "        0.33225387], dtype=float32),\n",
              " 3518: array([0.1530895 , 0.03454939, 0.06011408, ..., 0.29366952, 0.10783266,\n",
              "        0.06368589], dtype=float32),\n",
              " 3521: array([0.5608463 , 0.28342   , 0.81934506, ..., 0.35246712, 0.04677296,\n",
              "        0.05831887], dtype=float32),\n",
              " 3528: array([0.5614743 , 0.16739075, 0.6422275 , ..., 0.39434573, 0.2835456 ,\n",
              "        0.01077746], dtype=float32),\n",
              " 3532: array([0.6147269 , 0.29393345, 0.6420894 , ..., 0.00529731, 0.0030381 ,\n",
              "        0.0734286 ], dtype=float32),\n",
              " 3535: array([0.20392656, 0.28189093, 0.5260278 , ..., 1.42803   , 0.49834004,\n",
              "        0.4551112 ], dtype=float32),\n",
              " 3538: array([0.15304922, 0.09702232, 0.01960886, ..., 0.57513285, 0.51257336,\n",
              "        0.7888958 ], dtype=float32),\n",
              " 3579: array([0.19407834, 0.27221453, 0.3058453 , ..., 0.7024511 , 0.28556687,\n",
              "        0.11759423], dtype=float32),\n",
              " 3602: array([0.7063336 , 0.12536071, 0.330865  , ..., 1.0772963 , 0.19679682,\n",
              "        0.8843168 ], dtype=float32),\n",
              " 3613: array([0.09034179, 0.03316442, 2.374649  , ..., 0.4551961 , 0.25943303,\n",
              "        0.11182034], dtype=float32),\n",
              " 3623: array([0.3538436 , 0.7564483 , 0.09114226, ..., 0.1358766 , 0.80513823,\n",
              "        0.6405283 ], dtype=float32),\n",
              " 3628: array([0.68517035, 0.04663126, 0.95570356, ..., 0.28452635, 1.154255  ,\n",
              "        0.2513178 ], dtype=float32),\n",
              " 3637: array([0.47126395, 0.31072927, 0.5831381 , ..., 0.04483138, 0.45966238,\n",
              "        0.89683324], dtype=float32),\n",
              " 3640: array([0.14404757, 0.14834419, 0.14560896, ..., 0.06470425, 0.58928084,\n",
              "        0.40103197], dtype=float32),\n",
              " 3668: array([0.11094731, 0.2896039 , 0.03413151, ..., 0.7353739 , 0.40476555,\n",
              "        0.42927352], dtype=float32),\n",
              " 3671: array([0.21108356, 0.3246572 , 0.8595584 , ..., 0.12494096, 0.5705357 ,\n",
              "        1.5806746 ], dtype=float32),\n",
              " 3682: array([0.05948678, 0.25268653, 0.25572112, ..., 0.4837371 , 0.07863197,\n",
              "        0.21280757], dtype=float32),\n",
              " 3685: array([0.14022028, 0.15943465, 0.38035658, ..., 0.04750249, 0.5014258 ,\n",
              "        0.21239144], dtype=float32),\n",
              " 3692: array([0.35241693, 0.08138059, 0.08063583, ..., 0.01370452, 0.06969208,\n",
              "        1.658969  ], dtype=float32),\n",
              " 3713: array([0.5324614 , 0.38679892, 0.29173186, ..., 0.6232917 , 0.8447405 ,\n",
              "        0.6745836 ], dtype=float32),\n",
              " 3729: array([0.49439043, 0.14334261, 0.24495062, ..., 0.30291152, 0.1787434 ,\n",
              "        0.6845323 ], dtype=float32),\n",
              " 3735: array([0.32444727, 0.00990279, 0.47529417, ..., 0.3956408 , 0.38342342,\n",
              "        0.2787248 ], dtype=float32),\n",
              " 3737: array([0.27501655, 0.26953334, 0.4832284 , ..., 1.1109914 , 0.50590926,\n",
              "        0.7371296 ], dtype=float32),\n",
              " 3745: array([0.14985546, 0.54040176, 0.2696396 , ..., 0.24820778, 0.64194304,\n",
              "        0.29849696], dtype=float32),\n",
              " 3751: array([0.3276359 , 0.04543566, 0.61049676, ..., 0.01318783, 0.2329595 ,\n",
              "        0.02963483], dtype=float32),\n",
              " 3764: array([0.49913204, 0.27303243, 0.8991075 , ..., 0.78053975, 1.1304901 ,\n",
              "        0.48645577], dtype=float32),\n",
              " 3770: array([0.30864134, 0.03379522, 0.61191165, ..., 0.21780135, 0.7347428 ,\n",
              "        0.21842054], dtype=float32),\n",
              " 3782: array([0.31428027, 0.47777617, 0.61339   , ..., 0.03569967, 1.0372818 ,\n",
              "        0.08758917], dtype=float32),\n",
              " 3789: array([0.6984395 , 0.16734263, 0.644454  , ..., 0.12890017, 0.4957706 ,\n",
              "        0.5448201 ], dtype=float32),\n",
              " 3804: array([0.9631126 , 0.33828244, 0.822896  , ..., 0.20456803, 0.30858496,\n",
              "        0.16298036], dtype=float32),\n",
              " 3812: array([0.6477108 , 0.1869353 , 0.45682308, ..., 0.2675776 , 0.45169398,\n",
              "        0.31829596], dtype=float32),\n",
              " 3823: array([0.06308197, 0.36888355, 0.6968977 , ..., 1.0571957 , 0.27820635,\n",
              "        0.3265972 ], dtype=float32),\n",
              " 3827: array([0.41612875, 0.36633486, 0.74065596, ..., 0.09040426, 0.02172006,\n",
              "        1.0109496 ], dtype=float32),\n",
              " 3830: array([0.6465366 , 0.34287652, 0.76084346, ..., 1.7658638 , 0.10181282,\n",
              "        0.13070826], dtype=float32),\n",
              " 3860: array([0.58680505, 0.2840437 , 0.74551004, ..., 0.4070112 , 0.74698484,\n",
              "        0.42110234], dtype=float32),\n",
              " 3862: array([0.6501248 , 0.14619534, 0.14732482, ..., 0.33525074, 0.36433327,\n",
              "        0.13992518], dtype=float32),\n",
              " 3866: array([0.25754854, 0.22670057, 1.557755  , ..., 0.66174936, 0.7049483 ,\n",
              "        0.303581  ], dtype=float32),\n",
              " 3870: array([0.8509803 , 0.6929625 , 0.6525805 , ..., 0.10439792, 0.14381817,\n",
              "        0.9035989 ], dtype=float32),\n",
              " 3877: array([0.19435728, 0.11789078, 0.22981657, ..., 0.28995216, 0.43931767,\n",
              "        0.6864441 ], dtype=float32),\n",
              " 3897: array([0.44116372, 0.31662315, 0.13701443, ..., 0.12107012, 0.6701906 ,\n",
              "        0.77756613], dtype=float32),\n",
              " 3899: array([0.27367574, 0.16421816, 0.5087324 , ..., 0.0116643 , 1.2124709 ,\n",
              "        0.9122303 ], dtype=float32),\n",
              " 3911: array([0.04672066, 1.2087369 , 0.24270804, ..., 0.7012412 , 0.33781052,\n",
              "        0.1397911 ], dtype=float32),\n",
              " 3915: array([0.4354124 , 0.13313769, 1.822002  , ..., 0.29140586, 0.1453613 ,\n",
              "        0.3363411 ], dtype=float32),\n",
              " 3917: array([0.47686046, 0.72156364, 0.7501987 , ..., 0.08928615, 0.02326042,\n",
              "        0.04485707], dtype=float32),\n",
              " 3920: array([0.33244732, 0.16697012, 0.4848293 , ..., 0.509968  , 0.68484986,\n",
              "        0.38737196], dtype=float32),\n",
              " 3935: array([0.08986992, 0.2151891 , 0.04976952, ..., 0.3935752 , 0.37096697,\n",
              "        0.09237836], dtype=float32),\n",
              " 3941: array([0.43914178, 0.49191487, 0.346353  , ..., 0.75681293, 0.40964746,\n",
              "        0.6801287 ], dtype=float32),\n",
              " 3967: array([1.0066082 , 0.9490907 , 1.8594317 , ..., 0.5033774 , 1.235563  ,\n",
              "        0.06665688], dtype=float32),\n",
              " 3982: array([0.35810143, 0.29047376, 0.43467295, ..., 0.04370224, 0.02758999,\n",
              "        0.22402462], dtype=float32),\n",
              " 3988: array([0.3174599 , 0.2023288 , 0.65482503, ..., 0.19514701, 0.37582183,\n",
              "        0.556293  ], dtype=float32),\n",
              " 3992: array([0.17563875, 0.7012318 , 0.27715907, ..., 0.19959438, 0.271112  ,\n",
              "        0.24999274], dtype=float32),\n",
              " 3995: array([0.39544457, 0.4108085 , 0.24480286, ..., 0.14620462, 0.03428512,\n",
              "        0.256985  ], dtype=float32),\n",
              " 3999: array([0.44719172, 0.19697215, 0.16955489, ..., 0.74875027, 0.85283816,\n",
              "        0.12618451], dtype=float32),\n",
              " 4020: array([0.203806  , 0.11853239, 0.3192822 , ..., 0.24823892, 0.95166445,\n",
              "        0.2300353 ], dtype=float32),\n",
              " 4032: array([0.04570093, 0.36267245, 0.38099152, ..., 0.3387872 , 0.09559837,\n",
              "        0.9118168 ], dtype=float32),\n",
              " 4038: array([8.9034653e-01, 0.0000000e+00, 5.6911159e-01, ..., 4.7999442e-02,\n",
              "        7.3171360e-04, 1.5104905e-02], dtype=float32),\n",
              " 4042: array([0.12172455, 0.20890337, 0.18651381, ..., 0.36046243, 0.0606488 ,\n",
              "        0.08752884], dtype=float32),\n",
              " 4051: array([0.090633  , 0.45946938, 0.16342302, ..., 0.03254893, 0.5631114 ,\n",
              "        0.45535696], dtype=float32),\n",
              " 4057: array([0.03721876, 0.04521053, 0.77272934, ..., 0.07608154, 0.157293  ,\n",
              "        0.2317599 ], dtype=float32),\n",
              " 4065: array([0.7769021, 0.4963681, 1.3768456, ..., 0.6247163, 0.9371009,\n",
              "        1.7723551], dtype=float32),\n",
              " 4068: array([0.6709336 , 0.44538474, 0.47193787, ..., 0.58837056, 0.35214514,\n",
              "        0.13300283], dtype=float32),\n",
              " 4080: array([0.81231695, 0.21940666, 0.42829704, ..., 0.7505739 , 0.34973457,\n",
              "        1.0193074 ], dtype=float32),\n",
              " 4129: array([0.55277216, 0.07802518, 0.51726925, ..., 0.21922238, 0.43957174,\n",
              "        0.04541269], dtype=float32),\n",
              " 4130: array([0.18731357, 0.12260336, 0.25722617, ..., 0.40228888, 0.30338666,\n",
              "        0.26616156], dtype=float32),\n",
              " 4131: array([0.08534887, 0.20235194, 0.19102418, ..., 0.13455757, 0.04961933,\n",
              "        0.01668353], dtype=float32),\n",
              " 4132: array([0.14410517, 0.12234637, 0.58357096, ..., 1.230134  , 0.11695789,\n",
              "        0.65402997], dtype=float32),\n",
              " 4138: array([0.22752191, 0.12534046, 0.20213231, ..., 0.4100678 , 0.12234442,\n",
              "        0.17860188], dtype=float32),\n",
              " 4139: array([0.45241272, 0.19567375, 0.31283748, ..., 0.14551583, 0.06386419,\n",
              "        0.18548568], dtype=float32),\n",
              " 4140: array([0.1497018 , 0.05869074, 0.16864029, ..., 0.24693808, 1.0115055 ,\n",
              "        0.01112896], dtype=float32),\n",
              " 4159: array([0.3557884 , 0.48748928, 0.42016914, ..., 0.42725748, 0.06824543,\n",
              "        0.79716927], dtype=float32),\n",
              " 4172: array([0.200598  , 0.47294644, 0.34820962, ..., 0.28145468, 0.82845837,\n",
              "        0.52911574], dtype=float32),\n",
              " 4173: array([0.3907227 , 0.10602823, 0.25911433, ..., 0.06575659, 0.695451  ,\n",
              "        0.18376307], dtype=float32),\n",
              " 4180: array([0.41289216, 0.60605776, 0.2417082 , ..., 0.03603059, 0.36666346,\n",
              "        0.58004886], dtype=float32),\n",
              " 4189: array([0.00373265, 0.14862874, 0.25488842, ..., 0.06814864, 0.06183667,\n",
              "        0.3546998 ], dtype=float32),\n",
              " 4201: array([0.6151524 , 0.47397494, 0.06617172, ..., 0.15691884, 0.12448888,\n",
              "        0.10680766], dtype=float32),\n",
              " 4208: array([0.03391699, 0.22201848, 0.10259785, ..., 0.5567136 , 0.2097722 ,\n",
              "        0.01687481], dtype=float32),\n",
              " 4219: array([0.13592665, 0.25350884, 0.42576015, ..., 0.10789632, 0.28590864,\n",
              "        0.02802757], dtype=float32),\n",
              " 4239: array([0.28334326, 0.10524474, 0.03281484, ..., 0.69849014, 0.370241  ,\n",
              "        0.27463624], dtype=float32),\n",
              " 4244: array([0.9040321 , 0.22676225, 0.3468223 , ..., 0.01893022, 0.28608915,\n",
              "        0.03942231], dtype=float32),\n",
              " 4245: array([0.68428755, 0.2544275 , 0.37357116, ..., 0.5063683 , 0.50285065,\n",
              "        0.11342025], dtype=float32),\n",
              " 4259: array([0.8066273 , 0.38484764, 0.46569693, ..., 0.05020153, 0.42615736,\n",
              "        0.2869354 ], dtype=float32),\n",
              " 4260: array([0.03694928, 0.5288757 , 0.00502753, ..., 1.0517786 , 0.21282932,\n",
              "        0.19055474], dtype=float32),\n",
              " 4278: array([0.2832499 , 0.06091882, 0.01459115, ..., 0.1827685 , 0.23937768,\n",
              "        0.27392554], dtype=float32),\n",
              " 4282: array([0.57035345, 0.04648834, 0.56939054, ..., 1.0410341 , 1.0522611 ,\n",
              "        0.7833361 ], dtype=float32),\n",
              " 4289: array([0.74729806, 0.03828496, 0.3520618 , ..., 0.8264097 , 0.23849748,\n",
              "        0.6450895 ], dtype=float32),\n",
              " 4308: array([0.150233  , 0.18387344, 0.15387973, ..., 0.2451473 , 0.07570034,\n",
              "        0.69129837], dtype=float32),\n",
              " 4309: array([0.20993862, 0.17574212, 0.3633647 , ..., 0.5092921 , 1.0484674 ,\n",
              "        0.3440963 ], dtype=float32),\n",
              " 4318: array([0.84805894, 0.2460636 , 0.14551261, ..., 0.9310379 , 0.6260873 ,\n",
              "        0.01353033], dtype=float32),\n",
              " 4319: array([0.21214797, 0.6845459 , 0.4083476 , ..., 0.50076425, 0.08399495,\n",
              "        0.7727233 ], dtype=float32),\n",
              " 4322: array([0.28590328, 0.42118764, 0.14476383, ..., 0.0584807 , 0.13648903,\n",
              "        0.        ], dtype=float32),\n",
              " 4331: array([0.37715238, 0.2342756 , 0.47746205, ..., 0.6293663 , 0.1943205 ,\n",
              "        0.23381746], dtype=float32),\n",
              " 4360: array([0.2245966 , 0.11727738, 0.40226603, ..., 0.39291653, 0.18756458,\n",
              "        0.50445974], dtype=float32),\n",
              " 4376: array([0.2818084 , 0.14985332, 0.58840877, ..., 0.05973554, 0.6203213 ,\n",
              "        1.1675835 ], dtype=float32),\n",
              " 4377: array([0.2051797 , 0.7642478 , 1.1951323 , ..., 0.40011036, 0.21181428,\n",
              "        0.1455485 ], dtype=float32),\n",
              " 4385: array([0.33448675, 0.03598887, 0.0725669 , ..., 1.1894983 , 0.4620475 ,\n",
              "        0.20452115], dtype=float32),\n",
              " 4394: array([0.2183792 , 0.05379204, 0.10160629, ..., 0.44167906, 0.7339533 ,\n",
              "        0.73588943], dtype=float32),\n",
              " 4404: array([0.48204195, 0.11838901, 0.8952634 , ..., 0.75007015, 0.06973495,\n",
              "        0.0247836 ], dtype=float32),\n",
              " 4410: array([0.23073803, 0.4976189 , 0.02961876, ..., 0.35797292, 0.7077162 ,\n",
              "        0.31056216], dtype=float32),\n",
              " 4415: array([0.63066965, 0.6758052 , 0.794541  , ..., 0.11643249, 0.00983288,\n",
              "        0.5510398 ], dtype=float32),\n",
              " 4421: array([0.48782444, 0.23924738, 0.79797626, ..., 0.12581173, 0.44471395,\n",
              "        0.51044834], dtype=float32),\n",
              " 4424: array([0.3287372 , 0.31725246, 0.23354928, ..., 0.76072395, 0.3254944 ,\n",
              "        0.395986  ], dtype=float32),\n",
              " 4426: array([0.2611227 , 0.23660791, 0.23387006, ..., 0.03029846, 0.05291128,\n",
              "        0.02984513], dtype=float32),\n",
              " 4428: array([0.41147178, 0.6388252 , 0.26927674, ..., 0.00318714, 0.42168242,\n",
              "        0.2604851 ], dtype=float32),\n",
              " 4441: array([0.5208803 , 0.18709248, 0.19665956, ..., 0.2368201 , 0.02580181,\n",
              "        0.14811479], dtype=float32),\n",
              " 4442: array([0.1080973 , 0.19590485, 0.42016047, ..., 0.02561386, 0.1251759 ,\n",
              "        0.41304457], dtype=float32),\n",
              " 4444: array([1.8080835 , 0.14728415, 0.5428897 , ..., 0.08361152, 0.49193722,\n",
              "        0.78890944], dtype=float32),\n",
              " 4462: array([0.293705  , 0.17258161, 0.4603443 , ..., 0.04577761, 0.63272274,\n",
              "        0.74392664], dtype=float32),\n",
              " 4463: array([0.24429998, 0.603994  , 0.5586802 , ..., 0.4611134 , 0.24359825,\n",
              "        0.6501344 ], dtype=float32),\n",
              " 4471: array([0.11879271, 0.05197962, 0.10535365, ..., 0.30598488, 0.19188322,\n",
              "        0.2570538 ], dtype=float32),\n",
              " 4477: array([0.1605624 , 0.17440411, 0.10005558, ..., 0.14122672, 0.08177012,\n",
              "        0.5432992 ], dtype=float32),\n",
              " 4478: array([0.41809773, 0.37616032, 0.28765124, ..., 0.12778234, 0.        ,\n",
              "        0.37910557], dtype=float32),\n",
              " 4481: array([0.56336004, 0.03366476, 0.01869568, ..., 0.11075196, 0.8435333 ,\n",
              "        0.67025626], dtype=float32),\n",
              " 4488: array([0.04280641, 0.29426166, 1.6578492 , ..., 0.08133113, 0.6846757 ,\n",
              "        0.        ], dtype=float32),\n",
              " 4489: array([0.5046868 , 0.11258376, 0.35885435, ..., 0.5751837 , 0.4589365 ,\n",
              "        0.21558836], dtype=float32),\n",
              " 4490: array([0.0839505 , 0.31270543, 0.12110493, ..., 0.529955  , 0.05233863,\n",
              "        0.18488792], dtype=float32),\n",
              " 4502: array([0.1281138 , 0.48532885, 0.2453048 , ..., 0.38931584, 0.18530682,\n",
              "        0.5456749 ], dtype=float32),\n",
              " 4508: array([1.1467539 , 0.11171663, 0.34272453, ..., 0.25235683, 0.3314138 ,\n",
              "        0.09108785], dtype=float32),\n",
              " 4517: array([0.34068167, 0.7584051 , 0.3129298 , ..., 0.38872492, 0.3143291 ,\n",
              "        0.5432321 ], dtype=float32),\n",
              " 4527: array([0.11060289, 0.08561145, 0.5142559 , ..., 0.01461505, 0.3691988 ,\n",
              "        0.22853649], dtype=float32),\n",
              " 4535: array([0.8384869, 0.740232 , 0.7497839, ..., 0.6153725, 0.2824989,\n",
              "        0.1964981], dtype=float32),\n",
              " 4537: array([0.51437443, 0.30599988, 0.5385227 , ..., 0.12635216, 0.23660667,\n",
              "        0.35549837], dtype=float32),\n",
              " 4546: array([0.35317248, 0.25274307, 0.26652643, ..., 0.23941916, 0.29227743,\n",
              "        0.40752852], dtype=float32),\n",
              " 4549: array([0.2728731 , 0.03922083, 0.36103827, ..., 0.05414226, 0.04319768,\n",
              "        0.28000867], dtype=float32),\n",
              " 4555: array([0.23612386, 0.6407049 , 0.18035462, ..., 0.47175694, 0.20820954,\n",
              "        0.63425577], dtype=float32),\n",
              " 4567: array([0.14289832, 0.10342361, 0.66461337, ..., 0.3184359 , 0.7344332 ,\n",
              "        0.06262644], dtype=float32),\n",
              " 4571: array([0.12022264, 0.01954457, 0.12787952, ..., 0.00185822, 0.31874007,\n",
              "        0.13099378], dtype=float32),\n",
              " 4574: array([0.20083159, 0.08460967, 0.2644485 , ..., 0.06650683, 1.065824  ,\n",
              "        0.13397188], dtype=float32),\n",
              " 4575: array([0.06071982, 0.73603463, 0.8094885 , ..., 0.70733404, 0.32687116,\n",
              "        0.42055342], dtype=float32),\n",
              " 4578: array([0.5402144 , 0.03970613, 0.3191377 , ..., 0.00614459, 0.19001812,\n",
              "        1.1192057 ], dtype=float32),\n",
              " 4579: array([0.0188214 , 0.18399261, 0.7373829 , ..., 0.17359895, 0.4290632 ,\n",
              "        0.28788328], dtype=float32),\n",
              " 4587: array([0.7717184 , 0.44901708, 0.33999652, ..., 0.24693528, 0.04944085,\n",
              "        0.8836309 ], dtype=float32),\n",
              " 4595: array([0.18803678, 0.17734733, 0.18647788, ..., 0.19473033, 0.36397654,\n",
              "        0.47387505], dtype=float32),\n",
              " 4608: array([0.22942218, 0.25710034, 0.22554055, ..., 0.6605275 , 0.42063057,\n",
              "        0.04036798], dtype=float32),\n",
              " 4616: array([0.8721423 , 0.29275006, 1.1617316 , ..., 0.420058  , 1.2441803 ,\n",
              "        0.38438132], dtype=float32),\n",
              " 4622: array([0.43766308, 0.11703719, 0.18947256, ..., 0.        , 0.15426072,\n",
              "        0.4747681 ], dtype=float32),\n",
              " 4624: array([0.2635157 , 0.25349975, 0.68800735, ..., 0.3601762 , 0.14940746,\n",
              "        0.87708735], dtype=float32),\n",
              " 4642: array([0.11970643, 0.45354885, 0.03410885, ..., 0.5296544 , 0.49183148,\n",
              "        0.52849114], dtype=float32),\n",
              " 4647: array([0.39713222, 0.46099788, 0.14792864, ..., 0.03416013, 0.14779878,\n",
              "        0.15925182], dtype=float32),\n",
              " 4662: array([0.5015116 , 0.58431745, 0.24022135, ..., 0.19012819, 0.5331788 ,\n",
              "        0.20676109], dtype=float32),\n",
              " 4673: array([0.19795427, 1.2772832 , 0.37520045, ..., 0.56580913, 0.9544621 ,\n",
              "        0.23012179], dtype=float32),\n",
              " 4684: array([1.5585706, 0.4292159, 1.3826901, ..., 0.7581876, 1.1929629,\n",
              "        0.8852254], dtype=float32),\n",
              " 4694: array([0.14137346, 0.4180233 , 0.22453341, ..., 0.02008005, 0.20048498,\n",
              "        0.17904136], dtype=float32),\n",
              " 4702: array([0.12067363, 0.4477266 , 0.31751764, ..., 0.01851492, 0.5770565 ,\n",
              "        0.45806593], dtype=float32),\n",
              " 4704: array([0.09347798, 0.39253768, 0.28591055, ..., 0.07124192, 0.18786435,\n",
              "        0.26935607], dtype=float32),\n",
              " 4705: array([0.19278267, 0.02905973, 0.13040873, ..., 0.03305732, 0.        ,\n",
              "        0.0459714 ], dtype=float32),\n",
              " 4706: array([0.17654487, 0.3279927 , 0.17362496, ..., 0.14171356, 0.03328677,\n",
              "        0.41888356], dtype=float32),\n",
              " 4711: array([0.12130932, 0.16685691, 0.35993898, ..., 0.11058888, 0.1720907 ,\n",
              "        0.01912291], dtype=float32),\n",
              " 4714: array([0.4777497 , 0.22546381, 0.39543828, ..., 0.        , 0.4656238 ,\n",
              "        0.18205383], dtype=float32),\n",
              " 4716: array([0.20160931, 0.49376962, 0.2800094 , ..., 0.07846111, 0.        ,\n",
              "        0.01232968], dtype=float32),\n",
              " 4719: array([0.16149616, 0.32284057, 0.25227606, ..., 0.03725439, 0.5473288 ,\n",
              "        0.14263168], dtype=float32),\n",
              " 4739: array([0.28837204, 0.12958907, 0.24792874, ..., 0.22028176, 0.5657931 ,\n",
              "        0.40780264], dtype=float32),\n",
              " 4741: array([0.69766533, 0.20411521, 1.6292365 , ..., 0.6544342 , 0.18400592,\n",
              "        0.3046882 ], dtype=float32),\n",
              " 4761: array([0.06581921, 0.33491418, 1.4729532 , ..., 0.28022975, 0.5064285 ,\n",
              "        0.04547202], dtype=float32),\n",
              " 4762: array([0.09637571, 0.18849115, 0.48665488, ..., 0.04545926, 0.02495901,\n",
              "        0.50818276], dtype=float32),\n",
              " 4785: array([1.5051897 , 0.06327283, 0.5506103 , ..., 0.11483423, 0.        ,\n",
              "        0.13323031], dtype=float32),\n",
              " 4794: array([0.0611165 , 0.32708222, 0.22268376, ..., 0.43599147, 0.6557424 ,\n",
              "        0.02058292], dtype=float32),\n",
              " 4796: array([0.05390819, 0.39434278, 0.0471282 , ..., 0.50036246, 0.62472606,\n",
              "        0.05815779], dtype=float32),\n",
              " 4809: array([0.26191327, 0.41747916, 0.20151964, ..., 0.10823395, 0.24772102,\n",
              "        0.84122896], dtype=float32),\n",
              " 4820: array([0.44560677, 0.41563642, 0.42083296, ..., 0.07769942, 0.22255138,\n",
              "        1.2389063 ], dtype=float32),\n",
              " 4823: array([0.5094344 , 0.36166757, 0.9305265 , ..., 0.5190776 , 0.63620293,\n",
              "        0.34342518], dtype=float32),\n",
              " 4827: array([0.82132375, 0.8164253 , 0.09728055, ..., 0.02500957, 1.5532997 ,\n",
              "        0.5228189 ], dtype=float32),\n",
              " 4830: array([0.6374562 , 0.18448645, 0.2096868 , ..., 0.2135101 , 1.1871734 ,\n",
              "        0.8950078 ], dtype=float32),\n",
              " 4834: array([0.51218235, 0.22616814, 0.09187686, ..., 0.5475489 , 0.46845436,\n",
              "        0.9313447 ], dtype=float32),\n",
              " 4843: array([0.21795025, 0.32671958, 0.6392797 , ..., 0.40416825, 0.9070156 ,\n",
              "        0.11955466], dtype=float32),\n",
              " 4844: array([0.96384174, 0.5241974 , 0.30590728, ..., 0.5838031 , 0.18021785,\n",
              "        0.6004328 ], dtype=float32),\n",
              " 4859: array([0.36719787, 0.14970201, 0.07752462, ..., 0.47300172, 0.5831849 ,\n",
              "        0.35192198], dtype=float32),\n",
              " 4876: array([0.06013742, 1.3413018 , 0.2203952 , ..., 0.58669406, 0.16852197,\n",
              "        0.44908932], dtype=float32),\n",
              " 4880: array([0.47097728, 0.1797348 , 0.18162481, ..., 0.8753598 , 0.7354976 ,\n",
              "        0.2605822 ], dtype=float32),\n",
              " 4885: array([0.11392339, 0.15516102, 0.36487097, ..., 0.00667201, 0.47079226,\n",
              "        0.10509043], dtype=float32),\n",
              " 4888: array([0.265071  , 0.15397196, 0.23720527, ..., 0.47788203, 0.3398775 ,\n",
              "        0.4074968 ], dtype=float32),\n",
              " 4891: array([0.03624272, 0.5954883 , 0.44022742, ..., 0.06778483, 0.6822152 ,\n",
              "        0.61961746], dtype=float32),\n",
              " 4893: array([0.48605943, 0.22222827, 0.23798698, ..., 0.02387964, 0.06158583,\n",
              "        0.21595928], dtype=float32),\n",
              " 4901: array([0.45878997, 0.2532055 , 0.16564023, ..., 0.30385953, 0.6737554 ,\n",
              "        0.2172575 ], dtype=float32),\n",
              " 4903: array([1.4132776 , 0.59179217, 0.7054767 , ..., 0.06030336, 0.6325054 ,\n",
              "        0.05145463], dtype=float32),\n",
              " 4904: array([0.75280106, 0.36959392, 0.84288204, ..., 0.15293582, 0.66074824,\n",
              "        0.5534996 ], dtype=float32),\n",
              " 4920: array([0.15622827, 0.17514458, 0.5422691 , ..., 0.01199263, 0.36216027,\n",
              "        0.10274794], dtype=float32),\n",
              " 4931: array([0.38648292, 0.96999884, 0.53541994, ..., 0.1279485 , 0.31988487,\n",
              "        0.02399358], dtype=float32),\n",
              " 4947: array([0.41650397, 0.5085805 , 0.5335525 , ..., 0.12998196, 0.8297286 ,\n",
              "        0.4032746 ], dtype=float32),\n",
              " 4956: array([0.07423091, 0.7590108 , 0.02369764, ..., 0.4107957 , 0.72781336,\n",
              "        0.13071254], dtype=float32),\n",
              " 4963: array([0.37659204, 0.2133466 , 0.858237  , ..., 1.160023  , 0.58585215,\n",
              "        0.6719467 ], dtype=float32),\n",
              " 4968: array([0.29817978, 0.15583625, 0.24530736, ..., 0.66417444, 0.19832067,\n",
              "        0.1592323 ], dtype=float32),\n",
              " 4970: array([0.46865398, 0.17783274, 0.47189796, ..., 0.5417436 , 0.6044624 ,\n",
              "        0.6817384 ], dtype=float32),\n",
              " 4971: array([0.43425745, 0.1996359 , 0.49576145, ..., 0.51835287, 0.5049144 ,\n",
              "        0.4678378 ], dtype=float32),\n",
              " 4978: array([0.11402556, 0.09739454, 0.1949792 , ..., 0.53375065, 1.2200658 ,\n",
              "        0.7955074 ], dtype=float32),\n",
              " 4981: array([0.31476128, 0.24209172, 0.46400222, ..., 0.24283597, 0.979659  ,\n",
              "        0.4349446 ], dtype=float32),\n",
              " 4984: array([0.3761623 , 0.00638459, 0.43147075, ..., 0.00201356, 0.43484366,\n",
              "        0.06406674], dtype=float32),\n",
              " 4993: array([0.86368847, 0.040021  , 0.46470213, ..., 0.45325142, 0.4658574 ,\n",
              "        0.11419506], dtype=float32),\n",
              " 5005: array([0.51165134, 1.03587   , 0.26822096, ..., 0.3314128 , 1.4345356 ,\n",
              "        1.090342  ], dtype=float32),\n",
              " 5010: array([0.04510656, 0.08211638, 0.07901732, ..., 0.08596921, 0.3739645 ,\n",
              "        0.65016806], dtype=float32),\n",
              " 5011: array([0.7802906 , 0.9950851 , 0.35977888, ..., 0.15783624, 0.23323739,\n",
              "        0.41326195], dtype=float32),\n",
              " 5016: array([0.23243403, 0.09596138, 0.22977051, ..., 0.18863192, 0.55818975,\n",
              "        0.3166591 ], dtype=float32),\n",
              " 5018: array([0.11731447, 0.59345627, 0.12317012, ..., 0.16201842, 0.36058444,\n",
              "        0.00590073], dtype=float32),\n",
              " 5021: array([0.77408576, 0.32048213, 0.73591113, ..., 0.14127332, 1.6945735 ,\n",
              "        0.6764013 ], dtype=float32),\n",
              " 5028: array([0.05114349, 0.21427841, 0.47184175, ..., 0.58844805, 0.19831994,\n",
              "        0.1976913 ], dtype=float32),\n",
              " 5046: array([0.3532286 , 0.12064213, 0.8576542 , ..., 0.20485078, 0.12929443,\n",
              "        0.33145466], dtype=float32),\n",
              " 5073: array([2.3126853e-01, 3.2869905e-01, 5.4634082e-01, ..., 1.2538981e-02,\n",
              "        2.1413453e-02, 2.5377239e-04], dtype=float32),\n",
              " 5083: array([0.06122512, 0.10787959, 0.15180635, ..., 0.2810232 , 0.04961687,\n",
              "        0.4084096 ], dtype=float32),\n",
              " 5085: array([0.23807067, 0.08502123, 0.47899878, ..., 0.3539025 , 0.9075519 ,\n",
              "        0.01654641], dtype=float32),\n",
              " 5086: array([0.17267191, 0.29817653, 0.35794044, ..., 0.7407235 , 0.32627684,\n",
              "        0.8900352 ], dtype=float32),\n",
              " 5088: array([0.13797837, 0.21934189, 0.09138428, ..., 0.02208964, 0.02203782,\n",
              "        0.84583247], dtype=float32),\n",
              " 5094: array([0.20904541, 0.44280452, 0.65622985, ..., 0.23651809, 0.07981405,\n",
              "        0.22573587], dtype=float32),\n",
              " 5095: array([0.18218157, 0.25705945, 0.20121783, ..., 0.50720626, 0.06618316,\n",
              "        0.47796315], dtype=float32),\n",
              " 5099: array([0.1805264 , 0.15001279, 0.21993396, ..., 0.32797956, 0.28909996,\n",
              "        0.8817767 ], dtype=float32),\n",
              " 5111: array([0.22204441, 0.12020005, 0.16723461, ..., 0.05111131, 0.23084916,\n",
              "        0.14397141], dtype=float32),\n",
              " 5113: array([0.2114887 , 0.14733884, 0.49196613, ..., 0.17790586, 0.5857371 ,\n",
              "        0.05136028], dtype=float32),\n",
              " 5115: array([0.45277762, 0.4105816 , 0.20277208, ..., 0.10271922, 0.15533158,\n",
              "        0.15507196], dtype=float32),\n",
              " 5131: array([0.3198787 , 0.43206817, 0.45546174, ..., 0.6649948 , 0.43071997,\n",
              "        0.56998366], dtype=float32),\n",
              " 5139: array([0.14447883, 0.392968  , 0.44723684, ..., 0.35222873, 0.23343849,\n",
              "        0.06950499], dtype=float32),\n",
              " 5140: array([0.32209235, 0.18594605, 0.9088614 , ..., 1.3040618 , 0.15090668,\n",
              "        0.18897828], dtype=float32),\n",
              " 5142: array([0.45135695, 0.08233605, 0.2609517 , ..., 0.37047535, 0.8181957 ,\n",
              "        0.68178266], dtype=float32),\n",
              " 5151: array([0.35981047, 0.22387102, 0.22267869, ..., 0.14396933, 0.35819578,\n",
              "        0.7840271 ], dtype=float32),\n",
              " 5152: array([0.08993955, 0.30774194, 0.17969032, ..., 0.00300095, 0.85818666,\n",
              "        0.38648748], dtype=float32),\n",
              " 5156: array([0.25841874, 0.05966675, 0.13554461, ..., 0.24125671, 0.2722873 ,\n",
              "        0.7865571 ], dtype=float32),\n",
              " 5165: array([0.22859004, 0.46746248, 0.77709043, ..., 0.20064613, 0.54104346,\n",
              "        0.19403346], dtype=float32),\n",
              " 5169: array([0.3190241 , 0.14819366, 0.47665542, ..., 0.1170349 , 0.6316591 ,\n",
              "        0.6712011 ], dtype=float32),\n",
              " 5172: array([0.08445929, 0.45659202, 0.0781153 , ..., 0.02107467, 0.8959838 ,\n",
              "        0.9868098 ], dtype=float32),\n",
              " 5174: array([0.19677114, 0.11439453, 0.39127356, ..., 0.3390342 , 0.21370904,\n",
              "        0.48498797], dtype=float32),\n",
              " 5180: array([0.26918682, 0.655082  , 0.3821168 , ..., 1.3825772 , 0.6976857 ,\n",
              "        0.5724958 ], dtype=float32),\n",
              " 5198: array([0.17107233, 0.08389786, 0.68966085, ..., 0.10375904, 1.0234015 ,\n",
              "        0.20165664], dtype=float32),\n",
              " 5210: array([0.6560116 , 0.06019767, 0.6429728 , ..., 0.0781607 , 0.64280224,\n",
              "        0.7966627 ], dtype=float32),\n",
              " 5215: array([0.13071081, 0.291775  , 0.1333181 , ..., 0.18182316, 0.530955  ,\n",
              "        0.08107385], dtype=float32),\n",
              " 5219: array([0.50245696, 0.26006693, 0.8142803 , ..., 0.1435729 , 0.        ,\n",
              "        0.55248344], dtype=float32),\n",
              " 5237: array([0.06419618, 0.25055784, 0.05035486, ..., 0.09698366, 0.7239832 ,\n",
              "        0.4197893 ], dtype=float32),\n",
              " 5244: array([0.5518105 , 0.5633232 , 0.23151544, ..., 0.14585385, 0.21876106,\n",
              "        0.75005543], dtype=float32),\n",
              " 5253: array([0.3326884 , 0.25933868, 0.5440773 , ..., 1.116576  , 0.47283444,\n",
              "        0.30165988], dtype=float32),\n",
              " 5256: array([0.30352935, 0.7623286 , 0.38855398, ..., 0.29235303, 0.10846043,\n",
              "        0.022379  ], dtype=float32),\n",
              " 5259: array([0.19876118, 0.42341122, 0.15148497, ..., 0.05231107, 0.26619714,\n",
              "        1.2334791 ], dtype=float32),\n",
              " 5260: array([0.30804697, 0.26290587, 0.2436813 , ..., 0.37199274, 0.7197095 ,\n",
              "        0.3420344 ], dtype=float32),\n",
              " 5263: array([0.13177529, 0.32538518, 0.03986681, ..., 0.5140327 , 0.2112788 ,\n",
              "        0.13452777], dtype=float32),\n",
              " 5277: array([0.04050898, 0.08232211, 0.43383706, ..., 0.16029057, 1.1847436 ,\n",
              "        0.42722288], dtype=float32),\n",
              " 5288: array([0.3014874 , 0.14406736, 0.13805532, ..., 0.4657996 , 0.05356462,\n",
              "        0.04542842], dtype=float32),\n",
              " 5294: array([0.43089405, 0.24314608, 0.47219336, ..., 0.        , 0.02132007,\n",
              "        0.5292799 ], dtype=float32),\n",
              " 5303: array([0.28370756, 0.5478206 , 0.08982053, ..., 0.5373012 , 0.1868308 ,\n",
              "        0.18106449], dtype=float32),\n",
              " 5312: array([0.81177163, 0.3321364 , 0.86021054, ..., 0.31852996, 0.12857065,\n",
              "        0.42703336], dtype=float32),\n",
              " 5313: array([0.87542915, 0.16840419, 0.3158117 , ..., 0.14460847, 0.11947954,\n",
              "        0.3470574 ], dtype=float32),\n",
              " 5324: array([0.14686674, 0.06790981, 0.4407186 , ..., 0.291654  , 0.43659648,\n",
              "        0.33280593], dtype=float32),\n",
              " 5326: array([0.16125613, 0.12691848, 0.07172427, ..., 0.31264737, 0.03553145,\n",
              "        0.21600763], dtype=float32),\n",
              " 5335: array([0.00314691, 0.11588438, 0.24725823, ..., 0.1263544 , 0.19225141,\n",
              "        0.3448699 ], dtype=float32),\n",
              " 5336: array([0.7173308 , 0.777949  , 0.47535938, ..., 0.9476311 , 0.4353331 ,\n",
              "        0.2654764 ], dtype=float32),\n",
              " 5339: array([0.2274757 , 0.18845606, 0.12159856, ..., 0.39860764, 1.2365891 ,\n",
              "        0.14160892], dtype=float32),\n",
              " 5340: array([0.70294523, 0.47771797, 0.10754897, ..., 0.1649684 , 0.11151392,\n",
              "        0.11118579], dtype=float32),\n",
              " 5344: array([1.2638204 , 0.06552621, 0.10218364, ..., 0.88621616, 0.03577506,\n",
              "        0.04481234], dtype=float32),\n",
              " 5345: array([0.5086613 , 0.04844157, 0.415513  , ..., 0.2960875 , 0.22297283,\n",
              "        0.5072363 ], dtype=float32),\n",
              " 5355: array([0.30521455, 0.2686922 , 0.07149742, ..., 0.5662898 , 0.23570912,\n",
              "        0.41162947], dtype=float32),\n",
              " 5359: array([0.19418153, 0.07023429, 0.63739324, ..., 0.17251265, 0.10733042,\n",
              "        0.24831282], dtype=float32),\n",
              " 5360: array([0.09000909, 0.05120423, 0.13103786, ..., 0.5460109 , 0.31138942,\n",
              "        0.09705392], dtype=float32),\n",
              " 5362: array([0.9195077 , 0.10141182, 0.00622102, ..., 0.84051824, 0.11206126,\n",
              "        0.0504432 ], dtype=float32),\n",
              " 5368: array([0.1523456 , 0.1928975 , 0.08154938, ..., 0.60497135, 0.3453188 ,\n",
              "        0.4795788 ], dtype=float32),\n",
              " 5373: array([0.17914113, 0.3270853 , 0.22309425, ..., 0.38898188, 0.32522792,\n",
              "        0.4117229 ], dtype=float32),\n",
              " 5376: array([0.98049057, 0.21005607, 0.6245712 , ..., 0.37974852, 0.5038208 ,\n",
              "        1.1940789 ], dtype=float32),\n",
              " 5377: array([0.6120332 , 0.0565847 , 0.77140766, ..., 0.6867867 , 0.32848293,\n",
              "        0.1929976 ], dtype=float32),\n",
              " 5383: array([0.01984968, 0.5366636 , 0.15468067, ..., 0.34375775, 0.71142125,\n",
              "        0.6874764 ], dtype=float32),\n",
              " 5396: array([0.19186983, 0.11485973, 0.31184596, ..., 0.        , 0.07224335,\n",
              "        0.31897032], dtype=float32),\n",
              " 5424: array([0.21808545, 0.09460184, 0.13963304, ..., 0.1046204 , 1.0167642 ,\n",
              "        0.06794989], dtype=float32),\n",
              " 5425: array([0.43299025, 0.04250536, 0.33661705, ..., 0.1832663 , 1.1326835 ,\n",
              "        0.16970551], dtype=float32),\n",
              " 5430: array([0.3470754 , 0.27020264, 0.20764515, ..., 0.73881435, 0.27197406,\n",
              "        0.08267204], dtype=float32),\n",
              " 5434: array([0.17739739, 0.09572926, 0.96306056, ..., 1.5109434 , 0.54755116,\n",
              "        0.08969923], dtype=float32),\n",
              " 5435: array([0.32716304, 0.08359893, 0.22067654, ..., 0.12488166, 0.84689426,\n",
              "        0.15643424], dtype=float32),\n",
              " 5453: array([0.04973719, 0.6133053 , 0.93448913, ..., 1.037374  , 0.7687892 ,\n",
              "        0.82992446], dtype=float32),\n",
              " 5459: array([1.1381657 , 0.58649385, 0.36145568, ..., 0.573918  , 0.9979088 ,\n",
              "        0.15939823], dtype=float32),\n",
              " 5469: array([0.14243242, 0.1703941 , 2.0068243 , ..., 0.06455659, 0.08208171,\n",
              "        1.4312719 ], dtype=float32),\n",
              " 5471: array([0.3590414 , 0.10689794, 0.45057765, ..., 0.00433142, 0.94101596,\n",
              "        0.27407575], dtype=float32),\n",
              " 5472: array([0.15111437, 0.3939495 , 0.11947167, ..., 0.36728922, 0.9132743 ,\n",
              "        0.19816673], dtype=float32),\n",
              " 5482: array([1.2512283 , 0.20116058, 0.34472305, ..., 0.01510643, 1.8906355 ,\n",
              "        0.40519252], dtype=float32),\n",
              " 5483: array([0.35696214, 0.14559758, 0.32190278, ..., 0.00721937, 0.02509201,\n",
              "        0.8519905 ], dtype=float32),\n",
              " 5500: array([0.25785226, 0.37955144, 0.5054585 , ..., 1.3794291 , 0.08692501,\n",
              "        0.07341044], dtype=float32),\n",
              " 5505: array([0.92906815, 0.24705671, 1.1785271 , ..., 0.05187178, 0.38293245,\n",
              "        0.45162803], dtype=float32),\n",
              " 5508: array([0.63913846, 0.6509975 , 0.11263406, ..., 0.26210815, 0.9821745 ,\n",
              "        0.51575875], dtype=float32),\n",
              " 5522: array([0.40451816, 0.2155442 , 0.6724228 , ..., 0.09720973, 0.14461097,\n",
              "        1.0115275 ], dtype=float32),\n",
              " 5554: array([0.48736876, 0.28947186, 0.06213285, ..., 0.55956465, 0.49414927,\n",
              "        0.        ], dtype=float32),\n",
              " 5557: array([0.11201385, 0.1035204 , 0.23103848, ..., 0.0945807 , 0.04375273,\n",
              "        0.0795819 ], dtype=float32),\n",
              " 5559: array([0.18570946, 0.26877803, 0.1184625 , ..., 0.4832683 , 0.15013233,\n",
              "        0.86954784], dtype=float32),\n",
              " 5564: array([0.11157021, 0.47540396, 0.19812034, ..., 0.5417274 , 0.05929479,\n",
              "        0.2905763 ], dtype=float32),\n",
              " 5574: array([0.570487  , 0.46976852, 1.5018873 , ..., 0.6381233 , 0.7576857 ,\n",
              "        0.00535439], dtype=float32),\n",
              " 5587: array([0.49264848, 0.28282198, 0.21634738, ..., 0.06674655, 0.8585068 ,\n",
              "        0.7979855 ], dtype=float32),\n",
              " 5589: array([0.24437727, 0.30527437, 0.6600449 , ..., 1.0687423 , 0.52216077,\n",
              "        0.23751172], dtype=float32),\n",
              " 5608: array([0.13161741, 0.3349527 , 0.21900453, ..., 0.44412106, 0.11652763,\n",
              "        0.28730312], dtype=float32),\n",
              " 5612: array([0.5009075 , 0.7099881 , 0.33337924, ..., 0.13438274, 0.685508  ,\n",
              "        1.5753945 ], dtype=float32),\n",
              " 5614: array([0.12477175, 0.20233329, 0.1958522 , ..., 0.05586844, 0.43012357,\n",
              "        0.14135714], dtype=float32),\n",
              " 5615: array([0.53595936, 0.8272941 , 0.35981995, ..., 0.79648364, 0.7200424 ,\n",
              "        0.14888003], dtype=float32),\n",
              " 5619: array([0.20088772, 0.63967377, 0.270788  , ..., 0.5587463 , 0.11668249,\n",
              "        0.04959144], dtype=float32),\n",
              " 5620: array([0.09959911, 0.2687302 , 0.1315813 , ..., 0.7592912 , 0.03573502,\n",
              "        0.38282245], dtype=float32),\n",
              " 5632: array([0.0985695 , 0.07994471, 0.22105354, ..., 0.15480378, 0.28150737,\n",
              "        0.31320947], dtype=float32),\n",
              " 5638: array([0.21277879, 0.357639  , 0.27803856, ..., 0.62720895, 0.4403308 ,\n",
              "        0.46062386], dtype=float32),\n",
              " 5641: array([0.16032536, 0.18608011, 0.25471455, ..., 1.0036359 , 0.28044868,\n",
              "        0.348106  ], dtype=float32),\n",
              " 5643: array([0.33627337, 0.10475454, 0.43643928, ..., 0.41015667, 0.12034623,\n",
              "        0.26933277], dtype=float32),\n",
              " 5649: array([0.03401747, 0.13320637, 0.03868788, ..., 0.11520462, 0.03444448,\n",
              "        0.20760033], dtype=float32),\n",
              " 5667: array([0.3359092 , 0.8797678 , 0.85235953, ..., 0.96506906, 0.8024745 ,\n",
              "        0.21913266], dtype=float32),\n",
              " 5669: array([0.0926899 , 0.209272  , 0.51655275, ..., 0.8335236 , 0.77881813,\n",
              "        0.41878673], dtype=float32),\n",
              " 5678: array([0.20176557, 0.13989288, 0.41239846, ..., 0.00122064, 0.52149916,\n",
              "        0.48223263], dtype=float32),\n",
              " 5683: array([0.18503195, 0.06844866, 0.3206998 , ..., 0.3530577 , 0.93958056,\n",
              "        0.06500579], dtype=float32),\n",
              " 5684: array([0.13802752, 0.15507437, 0.36109555, ..., 0.57540226, 0.1474179 ,\n",
              "        0.09796765], dtype=float32),\n",
              " 5688: array([1.3420672 , 0.0335583 , 0.8277693 , ..., 1.363307  , 0.46318167,\n",
              "        0.3406952 ], dtype=float32),\n",
              " 5689: array([0.47750318, 0.20817336, 0.49152496, ..., 0.00117925, 1.1507179 ,\n",
              "        0.47894752], dtype=float32),\n",
              " 5692: array([0.05361177, 0.19887866, 0.15008464, ..., 0.42138177, 1.8567137 ,\n",
              "        0.17350307], dtype=float32),\n",
              " 5699: array([0.29793996, 0.14215924, 0.20640959, ..., 0.        , 0.99722564,\n",
              "        0.24619569], dtype=float32),\n",
              " 5700: array([0.59253633, 1.0295565 , 0.0575843 , ..., 0.20300883, 0.34195197,\n",
              "        0.0389765 ], dtype=float32),\n",
              " 5701: array([0.2623127 , 0.38506502, 0.12833312, ..., 0.17156881, 0.23952042,\n",
              "        0.30696517], dtype=float32),\n",
              " 5703: array([0.46688682, 0.03492067, 0.98534   , ..., 0.29255062, 0.7945649 ,\n",
              "        0.20334598], dtype=float32),\n",
              " 5715: array([0.41542265, 0.57978475, 1.439662  , ..., 0.58604825, 1.5087416 ,\n",
              "        0.08968328], dtype=float32),\n",
              " 5736: array([0.49242753, 0.21349843, 0.51220876, ..., 0.2819385 , 0.14140353,\n",
              "        0.20434512], dtype=float32),\n",
              " 5740: array([0.23474987, 0.09587854, 0.35430694, ..., 0.18264736, 0.06203633,\n",
              "        0.33595622], dtype=float32),\n",
              " 5745: array([0.07870445, 0.14213367, 0.05288625, ..., 0.16653106, 0.07249414,\n",
              "        0.47309712], dtype=float32),\n",
              " 5755: array([0.36559784, 0.21834056, 0.09473954, ..., 0.15418226, 0.03743207,\n",
              "        0.615561  ], dtype=float32),\n",
              " 5756: array([0.37238145, 0.3329792 , 0.37606972, ..., 0.22984746, 0.20243233,\n",
              "        0.18313637], dtype=float32),\n",
              " 5757: array([0.03000058, 0.29711413, 0.25062823, ..., 0.5954202 , 0.08710542,\n",
              "        0.6054847 ], dtype=float32),\n",
              " 5769: array([1.1279635 , 0.19696695, 0.2433871 , ..., 0.72902155, 0.19117032,\n",
              "        1.015097  ], dtype=float32),\n",
              " 5782: array([0.07444645, 0.00986915, 0.28666097, ..., 0.81551635, 1.3006141 ,\n",
              "        0.19975483], dtype=float32),\n",
              " 5785: array([0.3039957 , 0.27859896, 0.08002149, ..., 0.08528627, 0.07308044,\n",
              "        0.01865355], dtype=float32),\n",
              " 5809: array([0.2645373 , 0.16323043, 0.059356  , ..., 0.19705722, 0.30225718,\n",
              "        1.0866652 ], dtype=float32),\n",
              " 5811: array([0.11846092, 0.15288958, 0.25929713, ..., 0.63242805, 0.0884199 ,\n",
              "        0.45103294], dtype=float32),\n",
              " 5823: array([0.11056639, 0.12987143, 0.6147796 , ..., 0.4360157 , 0.01439534,\n",
              "        0.27139232], dtype=float32),\n",
              " 5828: array([0.21003595, 0.19492468, 0.23524258, ..., 0.45051023, 0.5344236 ,\n",
              "        0.16651578], dtype=float32),\n",
              " 5830: array([0.09253345, 0.07966068, 0.38327655, ..., 0.3412202 , 0.08332902,\n",
              "        0.18706395], dtype=float32),\n",
              " 5832: array([0.00327207, 0.23851374, 1.0212648 , ..., 0.7258859 , 0.00734912,\n",
              "        0.04397343], dtype=float32),\n",
              " 5862: array([0.16816197, 0.23689084, 0.16461347, ..., 0.04148184, 0.7848697 ,\n",
              "        0.45351624], dtype=float32),\n",
              " 5882: array([0.2280363 , 0.12928823, 0.20050335, ..., 0.10725629, 0.31804836,\n",
              "        0.3074446 ], dtype=float32),\n",
              " 5883: array([0.6691545 , 0.26262063, 0.7699307 , ..., 0.3711842 , 0.34051478,\n",
              "        0.23545514], dtype=float32),\n",
              " 5903: array([0.27776343, 0.1414005 , 0.0814186 , ..., 0.18212074, 0.2674585 ,\n",
              "        0.38022482], dtype=float32),\n",
              " 5906: array([0.39031887, 0.13920672, 0.4150337 , ..., 0.08680844, 0.04163421,\n",
              "        0.04785504], dtype=float32),\n",
              " 5907: array([0.38672215, 0.7417238 , 0.2579268 , ..., 0.12962514, 0.6502924 ,\n",
              "        1.1301142 ], dtype=float32),\n",
              " 5913: array([0.20826077, 0.1609667 , 0.11899515, ..., 0.19989324, 0.03598139,\n",
              "        0.05717599], dtype=float32),\n",
              " 5915: array([0.49213052, 0.07180002, 0.51408577, ..., 0.5484145 , 0.1262574 ,\n",
              "        0.00696699], dtype=float32),\n",
              " 5916: array([0.05470203, 0.48650524, 0.26549852, ..., 0.09936546, 0.1881501 ,\n",
              "        0.37755635], dtype=float32),\n",
              " 5917: array([0.48368052, 0.00315814, 0.3862063 , ..., 0.4462982 , 0.14703682,\n",
              "        0.        ], dtype=float32),\n",
              " 5933: array([0.02509852, 0.26116726, 0.38081595, ..., 0.88151467, 0.09686802,\n",
              "        0.27116358], dtype=float32),\n",
              " 5946: array([0.3639424 , 0.2025566 , 0.623352  , ..., 0.1489298 , 0.08477055,\n",
              "        0.28953633], dtype=float32),\n",
              " 5947: array([0.11668474, 0.18349744, 0.14690892, ..., 0.06203883, 0.16888916,\n",
              "        0.33276442], dtype=float32),\n",
              " 5962: array([0.48647645, 0.33089274, 0.3739027 , ..., 0.28227764, 0.6166711 ,\n",
              "        0.60282147], dtype=float32),\n",
              " 5967: array([0.0427747 , 0.1399759 , 0.2816831 , ..., 0.01806268, 0.06782417,\n",
              "        0.23501557], dtype=float32),\n",
              " 5991: array([0.32046378, 0.2003501 , 0.5284989 , ..., 0.23040855, 0.13328537,\n",
              "        0.36215174], dtype=float32),\n",
              " 5994: array([0.70195746, 0.02647706, 0.36089015, ..., 0.5359895 , 0.90405554,\n",
              "        0.34104484], dtype=float32),\n",
              " 6004: array([0.42619628, 0.13474607, 0.48519072, ..., 0.1484428 , 0.5037459 ,\n",
              "        0.8037675 ], dtype=float32),\n",
              " 6010: array([0.40962672, 0.56503344, 0.1253488 , ..., 0.31278923, 1.02458   ,\n",
              "        0.1861034 ], dtype=float32),\n",
              " 6016: array([0.1549794 , 0.16854054, 0.2940687 , ..., 0.01233167, 0.59533775,\n",
              "        0.0513359 ], dtype=float32),\n",
              " 6026: array([0.4327753 , 0.21686049, 0.329624  , ..., 0.19416356, 0.42422244,\n",
              "        0.1844314 ], dtype=float32),\n",
              " 6031: array([0.25247443, 0.70592904, 0.35565674, ..., 0.07305124, 0.0907944 ,\n",
              "        0.5588523 ], dtype=float32),\n",
              " 6041: array([0.37386096, 0.279522  , 0.29303083, ..., 0.6165969 , 0.14472152,\n",
              "        0.35390463], dtype=float32),\n",
              " 6042: array([0.07472074, 0.2977602 , 0.7130101 , ..., 0.18788345, 0.2133986 ,\n",
              "        0.6509755 ], dtype=float32),\n",
              " 6051: array([0.28779048, 0.47893214, 0.20720321, ..., 0.17078161, 0.4921886 ,\n",
              "        0.2150651 ], dtype=float32),\n",
              " 6053: array([0.20370756, 0.05975142, 0.10337113, ..., 1.1736565 , 0.07794625,\n",
              "        0.8267493 ], dtype=float32),\n",
              " 6057: array([0.0836311 , 0.1561219 , 0.2328833 , ..., 0.4732154 , 0.2939784 ,\n",
              "        0.37997258], dtype=float32),\n",
              " 6066: array([0.2859353 , 0.16433507, 0.5912283 , ..., 0.07715688, 0.10511611,\n",
              "        0.22019336], dtype=float32),\n",
              " 6068: array([0.591017  , 0.64570034, 0.04714179, ..., 0.1449574 , 0.2352089 ,\n",
              "        0.04788905], dtype=float32),\n",
              " 6075: array([1.5543352 , 0.3838849 , 0.23851521, ..., 0.8214315 , 0.32204837,\n",
              "        0.5144025 ], dtype=float32),\n",
              " 6101: array([0.2787121 , 0.51273394, 0.40829146, ..., 1.0641724 , 0.64482045,\n",
              "        0.82376695], dtype=float32),\n",
              " 6107: array([0.39581224, 0.51606774, 0.6777445 , ..., 0.15837917, 0.9515469 ,\n",
              "        0.16701722], dtype=float32),\n",
              " 6120: array([0.19293451, 0.12375124, 0.22430697, ..., 0.31478187, 0.18183929,\n",
              "        0.24456427], dtype=float32),\n",
              " 6140: array([1.1341965 , 0.12548214, 0.44671503, ..., 0.5915859 , 0.16073483,\n",
              "        0.7502879 ], dtype=float32),\n",
              " 6146: array([0.27806824, 0.21944681, 0.10771599, ..., 0.2958216 , 0.37251854,\n",
              "        0.3888425 ], dtype=float32),\n",
              " 6148: array([0.31594676, 0.23256305, 0.3604381 , ..., 0.01325228, 0.5122609 ,\n",
              "        0.54871285], dtype=float32),\n",
              " 6151: array([0.38092786, 0.6385169 , 0.46193415, ..., 0.579529  , 0.21665654,\n",
              "        0.11036753], dtype=float32),\n",
              " 6155: array([1.0049446 , 0.3017054 , 0.30912918, ..., 0.69610494, 0.29245389,\n",
              "        0.78299904], dtype=float32),\n",
              " 6160: array([0.95429635, 0.309133  , 0.17756817, ..., 0.5545181 , 0.37398544,\n",
              "        0.7910185 ], dtype=float32),\n",
              " 6178: array([0.20964979, 0.07719777, 0.08360265, ..., 0.06705304, 0.4622568 ,\n",
              "        1.0180027 ], dtype=float32),\n",
              " 6182: array([0.27280265, 0.3464707 , 0.43353385, ..., 0.5152464 , 0.04991969,\n",
              "        0.15106809], dtype=float32),\n",
              " 6190: array([0.5153388 , 0.27119178, 0.10772885, ..., 0.10127685, 0.32872254,\n",
              "        0.3195998 ], dtype=float32),\n",
              " 6197: array([0.37814766, 0.62855494, 0.20809996, ..., 0.15249316, 0.1814158 ,\n",
              "        0.75335526], dtype=float32),\n",
              " 6200: array([0.08873405, 0.01448601, 0.22817786, ..., 0.05533348, 0.43514526,\n",
              "        0.48972198], dtype=float32),\n",
              " 6216: array([0.31981283, 0.48557526, 0.43970048, ..., 1.3579093 , 0.01227147,\n",
              "        0.24198583], dtype=float32),\n",
              " 6225: array([0.02259564, 0.34290802, 0.21628788, ..., 0.34303045, 0.4646802 ,\n",
              "        0.7440027 ], dtype=float32),\n",
              " 6229: array([0.3249907 , 0.23430558, 0.12377769, ..., 0.13688529, 0.34670377,\n",
              "        1.0032871 ], dtype=float32),\n",
              " 6230: array([0.8857035 , 0.5806621 , 0.01797881, ..., 0.9376725 , 0.06932293,\n",
              "        0.694732  ], dtype=float32),\n",
              " 6233: array([0.24913739, 0.11111154, 0.79327494, ..., 0.57197714, 0.35472846,\n",
              "        0.31784368], dtype=float32),\n",
              " 6241: array([0.40622488, 0.08198608, 2.0193274 , ..., 0.34604812, 0.4147306 ,\n",
              "        0.20456776], dtype=float32),\n",
              " 6247: array([0.06925814, 0.1328969 , 0.4247173 , ..., 0.40193427, 0.01885751,\n",
              "        0.7919314 ], dtype=float32),\n",
              " 6253: array([0.20431048, 0.63070285, 0.12599486, ..., 0.2683948 , 0.27947468,\n",
              "        0.638309  ], dtype=float32),\n",
              " 6262: array([1.1951878e-01, 3.8702747e-01, 1.2494389e-01, ..., 1.6002255e-02,\n",
              "        1.6045303e-04, 5.0428450e-01], dtype=float32),\n",
              " 6263: array([0.08484015, 0.45383328, 1.8026267 , ..., 0.42831242, 0.7057003 ,\n",
              "        0.3408357 ], dtype=float32),\n",
              " 6268: array([0.14951485, 0.7508812 , 0.23056313, ..., 0.3408715 , 0.11312437,\n",
              "        0.41218802], dtype=float32),\n",
              " 6270: array([0.2016193 , 0.04679923, 0.20534654, ..., 0.50295544, 0.21156713,\n",
              "        0.6257256 ], dtype=float32),\n",
              " 6287: array([0.09416737, 0.2118354 , 0.31446528, ..., 0.10078546, 0.57746667,\n",
              "        0.759161  ], dtype=float32),\n",
              " 6293: array([0.46283987, 0.7132176 , 1.2690876 , ..., 1.3359009 , 0.25223267,\n",
              "        0.19973537], dtype=float32),\n",
              " 6295: array([1.2459867 , 0.19009861, 0.5886771 , ..., 0.        , 0.18626237,\n",
              "        1.0144253 ], dtype=float32),\n",
              " 6318: array([0.12674642, 0.26083624, 0.32229698, ..., 1.0704122 , 0.16440463,\n",
              "        0.63167214], dtype=float32),\n",
              " 6327: array([0.16382384, 0.0921092 , 0.37669525, ..., 0.4265958 , 0.83750856,\n",
              "        0.34523684], dtype=float32),\n",
              " 6332: array([1.0892687 , 0.5275219 , 0.83730733, ..., 0.78246796, 0.7901393 ,\n",
              "        0.60683465], dtype=float32),\n",
              " 6334: array([0.7377116 , 0.06793661, 0.2519297 , ..., 0.00625926, 0.79485434,\n",
              "        0.689536  ], dtype=float32),\n",
              " 6336: array([0.41079447, 0.6260958 , 0.6223191 , ..., 0.05252731, 0.55528855,\n",
              "        0.84584284], dtype=float32),\n",
              " 6338: array([0.27895957, 0.17786507, 0.296999  , ..., 0.64217204, 1.0386825 ,\n",
              "        0.76881754], dtype=float32),\n",
              " 6339: array([0.36036336, 0.45958138, 0.30156744, ..., 0.2391054 , 0.13498677,\n",
              "        0.1278517 ], dtype=float32),\n",
              " 6352: array([0.209758  , 1.0148679 , 0.15937701, ..., 0.33380827, 0.14742514,\n",
              "        0.3149226 ], dtype=float32),\n",
              " 6355: array([0.5380653 , 0.20471235, 0.46801984, ..., 0.        , 0.12048349,\n",
              "        0.338777  ], dtype=float32),\n",
              " 6357: array([0.03519383, 0.20908025, 0.40225902, ..., 0.00377041, 0.17750858,\n",
              "        0.3939456 ], dtype=float32),\n",
              " 6358: array([0.5243161 , 0.2913689 , 0.11803401, ..., 0.04046803, 0.22968985,\n",
              "        1.2201188 ], dtype=float32),\n",
              " 6363: array([0.2880907 , 0.49012217, 0.15081939, ..., 0.8052319 , 0.27256507,\n",
              "        1.1602728 ], dtype=float32),\n",
              " 6364: array([0.19719142, 0.40577313, 0.7026986 , ..., 0.84887993, 0.08049013,\n",
              "        0.4769554 ], dtype=float32),\n",
              " 6379: array([0.20336123, 0.21986759, 0.514614  , ..., 0.2152633 , 0.5368936 ,\n",
              "        0.3951609 ], dtype=float32),\n",
              " 6380: array([0.41659993, 0.12603469, 0.26296058, ..., 0.0050876 , 0.35462832,\n",
              "        0.14883667], dtype=float32),\n",
              " 6406: array([0.16668224, 0.45930776, 1.634491  , ..., 0.44811627, 0.7098074 ,\n",
              "        0.32415223], dtype=float32),\n",
              " 6407: array([0.71557426, 0.8664835 , 1.6582685 , ..., 0.2617359 , 0.8459413 ,\n",
              "        0.41754115], dtype=float32),\n",
              " 6409: array([0.08857831, 0.18347654, 0.8675766 , ..., 0.32471573, 0.7682475 ,\n",
              "        0.06076377], dtype=float32),\n",
              " 6414: array([0.20360757, 0.4002344 , 0.36499798, ..., 0.00958979, 0.20326373,\n",
              "        0.38306716], dtype=float32),\n",
              " 6421: array([0.23899353, 0.1326458 , 0.5514422 , ..., 0.25217354, 0.41389376,\n",
              "        0.66241825], dtype=float32),\n",
              " 6422: array([0.17048778, 0.03025317, 0.22592038, ..., 0.00114991, 0.40952432,\n",
              "        0.01227214], dtype=float32),\n",
              " 6424: array([0.45132324, 0.62071705, 1.7784746 , ..., 0.57069665, 0.7076531 ,\n",
              "        0.21962032], dtype=float32),\n",
              " 6426: array([0.1422018 , 0.20895827, 0.16776747, ..., 0.23252025, 0.2047559 ,\n",
              "        0.01573888], dtype=float32),\n",
              " 6428: array([0.08329935, 0.22940351, 0.9890036 , ..., 0.39059687, 0.07238963,\n",
              "        0.2310002 ], dtype=float32),\n",
              " 6432: array([0.19918324, 0.78329355, 0.3087324 , ..., 0.24202752, 0.4288662 ,\n",
              "        0.28065252], dtype=float32),\n",
              " 6447: array([0.2938469 , 0.3817583 , 0.1695487 , ..., 0.41677317, 0.06899629,\n",
              "        1.0972826 ], dtype=float32),\n",
              " 6451: array([0.62630254, 0.29726228, 0.3321883 , ..., 0.7371931 , 0.30332837,\n",
              "        0.43749163], dtype=float32),\n",
              " 6464: array([0.04561811, 0.33222246, 0.2128317 , ..., 0.2962783 , 0.18781066,\n",
              "        0.27210128], dtype=float32),\n",
              " 6465: array([0.69859457, 0.60175407, 0.29321623, ..., 0.41705158, 0.01132559,\n",
              "        0.91580147], dtype=float32),\n",
              " 6481: array([0.09008405, 0.10045743, 0.05720224, ..., 0.16500266, 0.02775623,\n",
              "        0.03731594], dtype=float32),\n",
              " 6488: array([0.10380353, 0.6941594 , 0.24067727, ..., 0.01079485, 0.08008829,\n",
              "        2.752287  ], dtype=float32),\n",
              " 6489: array([0.2760365 , 0.12975745, 0.08257826, ..., 0.04787157, 0.40471825,\n",
              "        0.76547354], dtype=float32),\n",
              " 6491: array([0.05653593, 0.4917646 , 0.9026653 , ..., 0.645777  , 0.12569383,\n",
              "        0.58074385], dtype=float32),\n",
              " 6512: array([0.01783072, 0.3937111 , 0.63192356, ..., 0.7410676 , 0.34425932,\n",
              "        0.3738226 ], dtype=float32),\n",
              " 6517: array([0.36438805, 0.19504446, 0.94040257, ..., 0.21133912, 0.23752551,\n",
              "        0.22218086], dtype=float32),\n",
              " 6518: array([0.3361903 , 0.42948985, 0.143376  , ..., 0.7404374 , 0.36110657,\n",
              "        0.2642953 ], dtype=float32),\n",
              " 6520: array([0.32559875, 0.2563703 , 0.3833514 , ..., 0.20731549, 0.7427584 ,\n",
              "        0.04529837], dtype=float32),\n",
              " 6522: array([0.85707045, 0.5559446 , 0.38294592, ..., 0.66344374, 0.36528796,\n",
              "        0.3416446 ], dtype=float32),\n",
              " 6531: array([0.5054482 , 0.44319642, 0.6421618 , ..., 0.4455792 , 0.6737511 ,\n",
              "        0.4105516 ], dtype=float32),\n",
              " 6539: array([0.4752492 , 0.30863136, 0.61271036, ..., 0.2201373 , 0.24925539,\n",
              "        0.28418696], dtype=float32),\n",
              " 6541: array([0.06682445, 0.08607447, 0.18832035, ..., 0.10419266, 0.06174803,\n",
              "        0.5126618 ], dtype=float32),\n",
              " 6560: array([0.03794298, 0.26888937, 0.09792478, ..., 0.38390374, 0.08098856,\n",
              "        0.14271018], dtype=float32),\n",
              " 6562: array([0.5763811 , 0.82433355, 0.27863127, ..., 0.39070743, 0.42784935,\n",
              "        0.3789058 ], dtype=float32),\n",
              " 6572: array([0.23033325, 0.12216666, 0.48736015, ..., 0.12557906, 0.21931353,\n",
              "        0.28400862], dtype=float32),\n",
              " 6578: array([0.55209714, 0.80026114, 0.5231768 , ..., 0.27288085, 0.82511616,\n",
              "        0.00772237], dtype=float32),\n",
              " 6586: array([0.21319641, 0.13925976, 0.1554809 , ..., 0.02080041, 0.21449572,\n",
              "        0.1907799 ], dtype=float32),\n",
              " 6590: array([0.23648082, 0.33454755, 0.47175696, ..., 0.8673006 , 0.23274647,\n",
              "        0.41618037], dtype=float32),\n",
              " 6595: array([0.07125377, 0.6754191 , 0.09616756, ..., 0.360447  , 0.6014548 ,\n",
              "        0.10913329], dtype=float32),\n",
              " 6599: array([0.9864799 , 0.17107622, 0.10837993, ..., 0.49546778, 0.0231651 ,\n",
              "        0.1916573 ], dtype=float32),\n",
              " 6602: array([1.0104547 , 0.11672215, 0.54687977, ..., 0.5279069 , 0.59303904,\n",
              "        0.42878366], dtype=float32),\n",
              " 6603: array([0.04567685, 0.15688966, 0.27319676, ..., 0.41855395, 0.6239792 ,\n",
              "        0.26523742], dtype=float32),\n",
              " 6627: array([0.4405201 , 0.14881547, 0.37288922, ..., 0.5719131 , 0.9697224 ,\n",
              "        0.4913757 ], dtype=float32),\n",
              " 6631: array([0.09654757, 0.09802634, 0.02896865, ..., 0.4706199 , 0.55869675,\n",
              "        0.13533512], dtype=float32),\n",
              " 6632: array([0.42731982, 0.9120586 , 0.1486001 , ..., 0.4084242 , 0.06632547,\n",
              "        0.44104695], dtype=float32),\n",
              " 6640: array([0.3773461 , 0.7632685 , 0.03911057, ..., 0.03668034, 0.41291803,\n",
              "        0.6475827 ], dtype=float32),\n",
              " 6647: array([0.24964416, 0.3806032 , 0.1826354 , ..., 0.2709502 , 0.16679694,\n",
              "        0.4339499 ], dtype=float32),\n",
              " 6651: array([0.25995922, 0.06363401, 0.50187606, ..., 0.20286298, 0.2295682 ,\n",
              "        1.1797702 ], dtype=float32),\n",
              " 6664: array([0.15835677, 0.21214443, 0.22264841, ..., 0.051631  , 0.0283924 ,\n",
              "        0.15605286], dtype=float32),\n",
              " 6675: array([0.63491356, 0.1750497 , 0.3792907 , ..., 0.9161124 , 0.715681  ,\n",
              "        0.190335  ], dtype=float32),\n",
              " 6692: array([0.45377752, 0.48335874, 0.37767243, ..., 0.1363815 , 0.15449017,\n",
              "        0.01082228], dtype=float32),\n",
              " 6709: array([0.19825122, 0.29978937, 1.0255768 , ..., 0.07785209, 0.39361346,\n",
              "        0.5403872 ], dtype=float32),\n",
              " 6710: array([0.37765363, 0.42062616, 0.45423526, ..., 0.40327227, 0.41510484,\n",
              "        0.58915675], dtype=float32),\n",
              " 6715: array([0.10499153, 0.42115405, 0.08390346, ..., 0.16261719, 1.1563631 ,\n",
              "        0.9331472 ], dtype=float32),\n",
              " 6721: array([0.10465272, 0.5275521 , 0.2288662 , ..., 0.0711058 , 0.06032427,\n",
              "        0.3477724 ], dtype=float32),\n",
              " 6725: array([0.03920837, 0.24685887, 0.5616994 , ..., 0.5510534 , 0.77256274,\n",
              "        0.07317212], dtype=float32),\n",
              " 6730: array([0.44610226, 0.3101542 , 0.8587479 , ..., 0.93722343, 0.0959719 ,\n",
              "        0.00285212], dtype=float32),\n",
              " 6733: array([0.1511586 , 0.46523854, 0.03373073, ..., 0.34407443, 0.01232701,\n",
              "        0.04560143], dtype=float32),\n",
              " 6744: array([0.75299644, 0.95669967, 0.32788095, ..., 0.07119782, 0.7094618 ,\n",
              "        0.02654032], dtype=float32),\n",
              " 6747: array([0.77954865, 0.68230134, 0.7137836 , ..., 1.1803086 , 0.86366355,\n",
              "        0.7695786 ], dtype=float32),\n",
              " 6749: array([0.23795952, 0.7058855 , 0.8128061 , ..., 0.08198045, 0.16308182,\n",
              "        0.13847122], dtype=float32),\n",
              " 6753: array([0.44722077, 0.19323725, 0.47024247, ..., 0.15363173, 0.19601862,\n",
              "        0.61113507], dtype=float32),\n",
              " 6760: array([0.21673551, 0.32907662, 0.17175166, ..., 0.8170829 , 0.01798448,\n",
              "        0.0572794 ], dtype=float32),\n",
              " 6764: array([0.24220648, 0.5233518 , 0.5232623 , ..., 0.24041674, 0.6137885 ,\n",
              "        0.32331932], dtype=float32),\n",
              " 6765: array([0.3027637 , 0.7320782 , 0.12292857, ..., 0.64970416, 0.5923146 ,\n",
              "        0.07451013], dtype=float32),\n",
              " 6773: array([0.10652042, 0.65484935, 0.48886642, ..., 0.36365625, 0.83466184,\n",
              "        0.5203699 ], dtype=float32),\n",
              " 6777: array([1.1226329 , 0.2642964 , 0.35646388, ..., 0.41840205, 0.00529246,\n",
              "        0.07847407], dtype=float32),\n",
              " 6780: array([0.33189723, 0.2726365 , 0.4665032 , ..., 0.01365253, 0.5125942 ,\n",
              "        0.59450984], dtype=float32),\n",
              " 6790: array([0.09205671, 0.01809324, 0.11193679, ..., 0.06292614, 0.665745  ,\n",
              "        0.318687  ], dtype=float32),\n",
              " 6792: array([0.26543832, 0.18328926, 0.36981243, ..., 0.2341072 , 0.61637175,\n",
              "        0.02567445], dtype=float32),\n",
              " 6800: array([0.16584012, 0.39266533, 0.40463373, ..., 0.10020915, 0.11487474,\n",
              "        0.5635348 ], dtype=float32),\n",
              " 6809: array([0.1290043 , 0.17491348, 0.2234801 , ..., 0.7847943 , 0.06557658,\n",
              "        0.20495167], dtype=float32),\n",
              " 6811: array([0.38515532, 0.28073645, 0.19545574, ..., 0.68087184, 0.5990995 ,\n",
              "        0.31030542], dtype=float32),\n",
              " 6819: array([0.14754072, 0.30501014, 0.05182783, ..., 0.00962663, 0.01214698,\n",
              "        0.14552806], dtype=float32),\n",
              " 6824: array([0.12974499, 0.0618123 , 0.16626489, ..., 0.15525149, 0.24141508,\n",
              "        0.7091446 ], dtype=float32),\n",
              " 6842: array([0.7415111 , 0.01576591, 1.2334596 , ..., 0.9905807 , 1.2687538 ,\n",
              "        0.37240672], dtype=float32),\n",
              " 6846: array([0.294008  , 0.09731386, 0.27025756, ..., 0.15501967, 0.46377298,\n",
              "        0.11310898], dtype=float32),\n",
              " 6860: array([0.00557689, 0.15660715, 0.24403751, ..., 0.37788254, 0.17183018,\n",
              "        0.0272722 ], dtype=float32),\n",
              " 6862: array([0.14909989, 0.13509306, 0.23768035, ..., 0.49172205, 0.7783778 ,\n",
              "        0.18567376], dtype=float32),\n",
              " 6873: array([0.08952047, 0.31818712, 0.29803902, ..., 0.51113397, 0.10935156,\n",
              "        0.10007349], dtype=float32),\n",
              " 6901: array([0.08305193, 0.5046252 , 0.0555175 , ..., 0.08517343, 0.23621514,\n",
              "        0.27136672], dtype=float32),\n",
              " 6914: array([0.06199179, 0.2639283 , 0.00747801, ..., 0.15754652, 0.10652605,\n",
              "        0.14549194], dtype=float32),\n",
              " 6920: array([0.17214918, 0.2117877 , 0.14159903, ..., 0.68682706, 0.6163382 ,\n",
              "        0.17016467], dtype=float32),\n",
              " 6935: array([0.3161957 , 0.5283169 , 0.6058841 , ..., 0.13671607, 0.7324537 ,\n",
              "        0.97025824], dtype=float32),\n",
              " 6936: array([0.39468837, 0.11849117, 0.2611305 , ..., 0.6221319 , 0.11763237,\n",
              "        1.0535942 ], dtype=float32),\n",
              " 6941: array([0.46274397, 0.43035585, 0.25351563, ..., 0.6625802 , 0.35832685,\n",
              "        0.26759285], dtype=float32),\n",
              " 6943: array([0.85354245, 0.04373152, 0.6959686 , ..., 0.44293135, 0.39299214,\n",
              "        1.4298077 ], dtype=float32),\n",
              " 6945: array([0.39434326, 0.40507087, 0.34055084, ..., 0.7997438 , 0.1763973 ,\n",
              "        0.33647397], dtype=float32),\n",
              " 6957: array([0.97301555, 0.27912164, 0.1040883 , ..., 0.8565489 , 0.06125144,\n",
              "        0.32513362], dtype=float32),\n",
              " 6964: array([1.4018803 , 0.28946322, 0.6168969 , ..., 0.3531007 , 0.9809615 ,\n",
              "        0.39340886], dtype=float32),\n",
              " 6973: array([0.18873619, 0.46151742, 0.4717027 , ..., 0.38573545, 0.7155566 ,\n",
              "        0.2211616 ], dtype=float32),\n",
              " 6981: array([0.6924665 , 0.13467605, 0.12071385, ..., 0.7289866 , 0.38633543,\n",
              "        0.5839628 ], dtype=float32),\n",
              " 6990: array([0.18419239, 0.7772341 , 0.15453403, ..., 0.12591791, 0.1543709 ,\n",
              "        0.16188836], dtype=float32),\n",
              " 6996: array([0.09808132, 0.45410436, 0.57952046, ..., 0.5146028 , 0.76904744,\n",
              "        0.11853432], dtype=float32),\n",
              " 6998: array([0.36461145, 0.46543813, 0.2943166 , ..., 1.3342501 , 1.1746101 ,\n",
              "        0.29556674], dtype=float32),\n",
              " 7022: array([0.41546717, 0.22877783, 0.39526612, ..., 0.33215114, 0.01388186,\n",
              "        0.21852158], dtype=float32),\n",
              " 7028: array([0.04622727, 0.69034386, 0.07881536, ..., 0.9594552 , 0.48805815,\n",
              "        0.08334816], dtype=float32),\n",
              " 7035: array([0.4854309 , 0.13297977, 0.19212309, ..., 0.00265252, 0.39997962,\n",
              "        0.32677716], dtype=float32),\n",
              " 7040: array([0.09546094, 0.48246458, 0.15433437, ..., 0.01404781, 0.18445939,\n",
              "        0.289627  ], dtype=float32),\n",
              " 7048: array([0.5399591 , 0.07840495, 0.07744814, ..., 0.02829513, 0.27795568,\n",
              "        1.3371952 ], dtype=float32),\n",
              " 7049: array([0.26365033, 0.02857426, 0.75911367, ..., 0.92044234, 0.18522686,\n",
              "        0.15687883], dtype=float32),\n",
              " 7069: array([0.2672408 , 0.13986243, 0.06125494, ..., 0.39435768, 0.22555459,\n",
              "        0.03756395], dtype=float32),\n",
              " 7090: array([0.04960013, 0.5726049 , 0.9145023 , ..., 1.0978088 , 0.28513828,\n",
              "        0.2731118 ], dtype=float32),\n",
              " 7095: array([0.5886839 , 0.27482685, 1.0772027 , ..., 0.35729694, 0.3484726 ,\n",
              "        0.3259336 ], dtype=float32),\n",
              " 7103: array([0.15898101, 0.13101639, 0.8893982 , ..., 1.1257391 , 0.2911672 ,\n",
              "        0.3863265 ], dtype=float32),\n",
              " 7104: array([0.6878516 , 0.63263   , 1.2369732 , ..., 0.43594462, 0.75207806,\n",
              "        0.06879367], dtype=float32),\n",
              " 7116: array([0.31951916, 0.04664945, 0.3038342 , ..., 0.06040768, 0.40231162,\n",
              "        0.17563339], dtype=float32),\n",
              " 7123: array([0.02642467, 0.02320473, 0.37585893, ..., 0.19725934, 0.69225836,\n",
              "        0.04716469], dtype=float32),\n",
              " 7124: array([3.0960608e-01, 3.7378708e-01, 2.8993577e-01, ..., 3.3001822e-01,\n",
              "        1.6282058e-04, 1.5682667e-01], dtype=float32),\n",
              " 7129: array([0.10452152, 0.21647604, 0.4558064 , ..., 0.25344   , 0.49587256,\n",
              "        0.5148975 ], dtype=float32),\n",
              " 7139: array([0.08236261, 0.23439474, 0.22878923, ..., 0.11484655, 0.12364633,\n",
              "        0.18990523], dtype=float32),\n",
              " 7143: array([0.30642977, 0.07634667, 0.43460158, ..., 0.35555875, 0.35183525,\n",
              "        0.35045508], dtype=float32),\n",
              " 7145: array([0.00137607, 0.42923176, 0.05439037, ..., 0.68281657, 0.40587124,\n",
              "        0.01451605], dtype=float32),\n",
              " 7150: array([0.287425  , 0.14787754, 0.19009362, ..., 0.70006335, 0.5101577 ,\n",
              "        0.92381287], dtype=float32),\n",
              " 7159: array([0.13669223, 0.18483478, 0.17527355, ..., 0.49599677, 0.5215479 ,\n",
              "        0.54765093], dtype=float32),\n",
              " 7167: array([0.10251816, 0.4189058 , 0.2178408 , ..., 0.92492217, 0.23239347,\n",
              "        0.42096853], dtype=float32),\n",
              " 7174: array([0.3206345 , 0.33826768, 0.38975593, ..., 0.7380825 , 0.30773607,\n",
              "        0.18748891], dtype=float32),\n",
              " 7179: array([0.26765513, 0.11245212, 1.6951838 , ..., 0.06016455, 0.05949409,\n",
              "        0.86208403], dtype=float32),\n",
              " 7201: array([0.08151832, 0.05562238, 0.20091838, ..., 0.5888766 , 0.4948325 ,\n",
              "        0.2742375 ], dtype=float32),\n",
              " 7205: array([0.20170346, 0.30156857, 0.10903881, ..., 0.16651556, 0.21598786,\n",
              "        0.44846338], dtype=float32),\n",
              " 7220: array([0.09689704, 0.4289102 , 1.1188619 , ..., 0.9402724 , 0.17924717,\n",
              "        0.74052864], dtype=float32),\n",
              " 7221: array([0.09343278, 0.3279331 , 0.06544951, ..., 0.00973857, 0.01995533,\n",
              "        0.00882936], dtype=float32),\n",
              " 7224: array([0.11604907, 0.23084486, 0.14819258, ..., 0.5246276 , 0.24512465,\n",
              "        0.32660267], dtype=float32),\n",
              " 7228: array([0.43070084, 0.48570207, 0.8923459 , ..., 0.634076  , 0.29503933,\n",
              "        0.34544337], dtype=float32),\n",
              " 7232: array([0.6602703 , 0.25406262, 0.0829652 , ..., 0.08667644, 0.20830445,\n",
              "        0.06426428], dtype=float32),\n",
              " 7239: array([0.2374032 , 0.22053845, 0.5490297 , ..., 0.5616111 , 0.4821132 ,\n",
              "        0.18044367], dtype=float32),\n",
              " 7247: array([0.35003957, 0.14102721, 0.31396487, ..., 0.30810308, 0.46320903,\n",
              "        0.22753724], dtype=float32),\n",
              " 7251: array([0.19593294, 0.22740126, 0.56394804, ..., 0.30038458, 0.11457931,\n",
              "        0.03738138], dtype=float32),\n",
              " 7275: array([0.20999163, 0.1555399 , 0.23425463, ..., 0.8133822 , 0.08202979,\n",
              "        0.28115454], dtype=float32),\n",
              " 7277: array([0.9186678 , 0.54986656, 0.4195513 , ..., 0.4262775 , 0.4500359 ,\n",
              "        0.23193324], dtype=float32),\n",
              " 7307: array([0.11520432, 0.5937845 , 0.29744327, ..., 0.3075819 , 0.05044843,\n",
              "        0.1960723 ], dtype=float32),\n",
              " 7318: array([0.7307949 , 0.5066299 , 0.5649085 , ..., 0.72243464, 0.9611766 ,\n",
              "        0.4107349 ], dtype=float32),\n",
              " 7319: array([0.72746694, 0.25660717, 0.19789366, ..., 0.90910125, 1.1519732 ,\n",
              "        0.8415247 ], dtype=float32),\n",
              " 7357: array([0.23808384, 0.01659432, 0.56570697, ..., 0.21145608, 0.51880515,\n",
              "        0.24635753], dtype=float32),\n",
              " 7361: array([0.78376925, 0.00743188, 0.26286045, ..., 0.02577328, 0.02183322,\n",
              "        1.6131334 ], dtype=float32),\n",
              " 7367: array([0.42261797, 0.41848683, 0.18472366, ..., 0.4696527 , 0.28602618,\n",
              "        0.56399584], dtype=float32),\n",
              " 7393: array([0.1624509 , 0.4525937 , 0.1253785 , ..., 0.00630814, 0.13106869,\n",
              "        0.60905486], dtype=float32),\n",
              " 7396: array([0.8472405 , 0.25810435, 0.0872637 , ..., 0.19078487, 0.88117534,\n",
              "        0.02159526], dtype=float32),\n",
              " 7420: array([1.0343957 , 0.659742  , 3.0325875 , ..., 0.19124927, 1.0625131 ,\n",
              "        1.1465782 ], dtype=float32),\n",
              " 7424: array([0.43774736, 0.17631334, 0.16789529, ..., 0.05309806, 0.09522401,\n",
              "        0.1383628 ], dtype=float32),\n",
              " 7452: array([0.3209011 , 0.14492817, 0.34037364, ..., 0.14737482, 0.4918328 ,\n",
              "        0.6280268 ], dtype=float32),\n",
              " 7455: array([0.14815196, 0.20036212, 0.35116422, ..., 0.3824445 , 0.33377376,\n",
              "        0.7863271 ], dtype=float32),\n",
              " 7476: array([0.13192871, 0.04544699, 0.06424425, ..., 0.2428301 , 0.14323562,\n",
              "        0.2529628 ], dtype=float32),\n",
              " 7489: array([0.01887824, 0.12323055, 0.12127557, ..., 0.04720048, 0.02135892,\n",
              "        0.47135922], dtype=float32),\n",
              " 7498: array([0.2192733 , 0.3896076 , 0.03193193, ..., 0.5972465 , 0.4022809 ,\n",
              "        0.03836539], dtype=float32),\n",
              " 7500: array([0.46744728, 0.21527754, 1.1709975 , ..., 0.24275348, 1.3983301 ,\n",
              "        0.49985677], dtype=float32),\n",
              " 7503: array([0.33932662, 0.08363305, 0.15081027, ..., 0.53432053, 0.58253264,\n",
              "        0.4729678 ], dtype=float32),\n",
              " 7504: array([0.19617966, 0.11827847, 0.4560017 , ..., 0.10305257, 0.10804674,\n",
              "        0.14321442], dtype=float32),\n",
              " 7510: array([0.23802102, 0.62661994, 0.19964525, ..., 0.02209453, 0.05789552,\n",
              "        0.00894232], dtype=float32),\n",
              " ...}"
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#train_imgs_features{7224}"
      ],
      "metadata": {
        "id": "XKK4vAzJkkdy"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(train_imgs_features.values)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dr-csnbO_OvX",
        "outputId": "29d40c65-a30b-4faf-fb60-f7936c5016d3"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<built-in method values of dict object at 0x7fc2eb29aa40>\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "list(train_imgs_features.keys())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "em2tkIqFAvOR",
        "outputId": "94d391a0-57a6-4b70-968e-0ab4aee6a04d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[9,\n",
              " 25,\n",
              " 30,\n",
              " 34,\n",
              " 36,\n",
              " 49,\n",
              " 61,\n",
              " 64,\n",
              " 71,\n",
              " 72,\n",
              " 77,\n",
              " 78,\n",
              " 81,\n",
              " 86,\n",
              " 89,\n",
              " 92,\n",
              " 94,\n",
              " 109,\n",
              " 110,\n",
              " 113,\n",
              " 127,\n",
              " 138,\n",
              " 142,\n",
              " 144,\n",
              " 149,\n",
              " 151,\n",
              " 154,\n",
              " 165,\n",
              " 194,\n",
              " 201,\n",
              " 247,\n",
              " 250,\n",
              " 260,\n",
              " 263,\n",
              " 307,\n",
              " 308,\n",
              " 309,\n",
              " 312,\n",
              " 315,\n",
              " 321,\n",
              " 322,\n",
              " 326,\n",
              " 332,\n",
              " 349,\n",
              " 368,\n",
              " 370,\n",
              " 382,\n",
              " 384,\n",
              " 389,\n",
              " 394,\n",
              " 404,\n",
              " 419,\n",
              " 431,\n",
              " 436,\n",
              " 438,\n",
              " 443,\n",
              " 446,\n",
              " 450,\n",
              " 471,\n",
              " 490,\n",
              " 491,\n",
              " 508,\n",
              " 510,\n",
              " 514,\n",
              " 529,\n",
              " 531,\n",
              " 532,\n",
              " 540,\n",
              " 542,\n",
              " 560,\n",
              " 562,\n",
              " 572,\n",
              " 575,\n",
              " 581,\n",
              " 584,\n",
              " 595,\n",
              " 597,\n",
              " 605,\n",
              " 612,\n",
              " 620,\n",
              " 625,\n",
              " 629,\n",
              " 634,\n",
              " 643,\n",
              " 650,\n",
              " 656,\n",
              " 659,\n",
              " 670,\n",
              " 671,\n",
              " 673,\n",
              " 681,\n",
              " 684,\n",
              " 690,\n",
              " 706,\n",
              " 714,\n",
              " 716,\n",
              " 722,\n",
              " 723,\n",
              " 731,\n",
              " 735,\n",
              " 753,\n",
              " 754,\n",
              " 762,\n",
              " 781,\n",
              " 790,\n",
              " 795,\n",
              " 797,\n",
              " 801,\n",
              " 813,\n",
              " 821,\n",
              " 825,\n",
              " 828,\n",
              " 839,\n",
              " 853,\n",
              " 882,\n",
              " 897,\n",
              " 901,\n",
              " 902,\n",
              " 908,\n",
              " 909,\n",
              " 913,\n",
              " 925,\n",
              " 927,\n",
              " 934,\n",
              " 941,\n",
              " 943,\n",
              " 955,\n",
              " 960,\n",
              " 965,\n",
              " 977,\n",
              " 982,\n",
              " 984,\n",
              " 996,\n",
              " 1006,\n",
              " 1011,\n",
              " 1014,\n",
              " 1025,\n",
              " 1036,\n",
              " 1053,\n",
              " 1059,\n",
              " 1072,\n",
              " 1084,\n",
              " 1085,\n",
              " 1090,\n",
              " 1098,\n",
              " 1099,\n",
              " 1102,\n",
              " 1107,\n",
              " 1108,\n",
              " 1111,\n",
              " 1122,\n",
              " 1139,\n",
              " 1144,\n",
              " 1145,\n",
              " 1155,\n",
              " 1166,\n",
              " 1168,\n",
              " 1183,\n",
              " 1200,\n",
              " 1204,\n",
              " 1213,\n",
              " 1216,\n",
              " 1224,\n",
              " 1232,\n",
              " 1237,\n",
              " 1238,\n",
              " 1261,\n",
              " 1264,\n",
              " 1271,\n",
              " 1282,\n",
              " 1295,\n",
              " 1298,\n",
              " 1306,\n",
              " 1307,\n",
              " 1308,\n",
              " 1311,\n",
              " 1315,\n",
              " 1319,\n",
              " 1323,\n",
              " 1330,\n",
              " 1332,\n",
              " 1350,\n",
              " 1355,\n",
              " 1359,\n",
              " 1360,\n",
              " 1366,\n",
              " 1375,\n",
              " 1381,\n",
              " 1386,\n",
              " 1390,\n",
              " 1392,\n",
              " 1397,\n",
              " 1401,\n",
              " 1403,\n",
              " 1407,\n",
              " 1408,\n",
              " 1424,\n",
              " 1431,\n",
              " 1451,\n",
              " 1453,\n",
              " 1455,\n",
              " 1472,\n",
              " 1488,\n",
              " 1496,\n",
              " 1497,\n",
              " 1501,\n",
              " 1505,\n",
              " 1507,\n",
              " 1510,\n",
              " 1515,\n",
              " 1518,\n",
              " 1522,\n",
              " 1523,\n",
              " 1526,\n",
              " 1527,\n",
              " 1536,\n",
              " 1548,\n",
              " 1558,\n",
              " 1562,\n",
              " 1569,\n",
              " 1579,\n",
              " 1580,\n",
              " 1586,\n",
              " 1589,\n",
              " 1596,\n",
              " 1611,\n",
              " 1622,\n",
              " 1625,\n",
              " 1637,\n",
              " 1639,\n",
              " 1645,\n",
              " 1670,\n",
              " 1674,\n",
              " 1681,\n",
              " 1688,\n",
              " 1697,\n",
              " 1706,\n",
              " 1709,\n",
              " 1712,\n",
              " 1720,\n",
              " 1732,\n",
              " 1737,\n",
              " 1756,\n",
              " 1762,\n",
              " 1764,\n",
              " 1771,\n",
              " 1774,\n",
              " 1777,\n",
              " 1781,\n",
              " 1785,\n",
              " 1786,\n",
              " 1790,\n",
              " 1792,\n",
              " 1804,\n",
              " 1810,\n",
              " 1811,\n",
              " 1813,\n",
              " 1815,\n",
              " 1822,\n",
              " 1837,\n",
              " 1864,\n",
              " 1875,\n",
              " 1877,\n",
              " 1888,\n",
              " 1895,\n",
              " 1900,\n",
              " 1902,\n",
              " 1906,\n",
              " 1907,\n",
              " 1911,\n",
              " 1912,\n",
              " 1915,\n",
              " 1924,\n",
              " 1926,\n",
              " 1941,\n",
              " 1942,\n",
              " 1943,\n",
              " 1947,\n",
              " 1958,\n",
              " 1966,\n",
              " 1994,\n",
              " 1997,\n",
              " 1999,\n",
              " 2001,\n",
              " 2007,\n",
              " 2024,\n",
              " 2055,\n",
              " 2056,\n",
              " 2066,\n",
              " 2068,\n",
              " 2072,\n",
              " 2083,\n",
              " 2089,\n",
              " 2093,\n",
              " 2106,\n",
              " 2114,\n",
              " 2135,\n",
              " 2148,\n",
              " 2150,\n",
              " 2178,\n",
              " 2184,\n",
              " 2193,\n",
              " 2197,\n",
              " 2209,\n",
              " 2211,\n",
              " 2217,\n",
              " 2229,\n",
              " 2232,\n",
              " 2244,\n",
              " 2258,\n",
              " 2270,\n",
              " 2276,\n",
              " 2278,\n",
              " 2279,\n",
              " 2280,\n",
              " 2281,\n",
              " 2283,\n",
              " 2284,\n",
              " 2296,\n",
              " 2309,\n",
              " 2337,\n",
              " 2342,\n",
              " 2347,\n",
              " 2349,\n",
              " 2369,\n",
              " 2372,\n",
              " 2374,\n",
              " 2377,\n",
              " 2389,\n",
              " 2400,\n",
              " 2402,\n",
              " 2411,\n",
              " 2415,\n",
              " 2429,\n",
              " 2444,\n",
              " 2445,\n",
              " 2446,\n",
              " 2448,\n",
              " 2451,\n",
              " 2459,\n",
              " 2466,\n",
              " 2470,\n",
              " 2471,\n",
              " 2496,\n",
              " 2498,\n",
              " 2531,\n",
              " 2536,\n",
              " 2543,\n",
              " 2544,\n",
              " 2545,\n",
              " 2555,\n",
              " 2559,\n",
              " 2560,\n",
              " 2563,\n",
              " 2567,\n",
              " 2570,\n",
              " 2575,\n",
              " 2583,\n",
              " 2585,\n",
              " 2591,\n",
              " 2602,\n",
              " 2606,\n",
              " 2608,\n",
              " 2614,\n",
              " 2618,\n",
              " 2619,\n",
              " 2623,\n",
              " 2624,\n",
              " 2639,\n",
              " 2644,\n",
              " 2645,\n",
              " 2658,\n",
              " 2664,\n",
              " 2672,\n",
              " 2686,\n",
              " 2687,\n",
              " 2691,\n",
              " 2693,\n",
              " 2697,\n",
              " 2703,\n",
              " 2732,\n",
              " 2742,\n",
              " 2752,\n",
              " 2754,\n",
              " 2755,\n",
              " 2758,\n",
              " 2770,\n",
              " 2774,\n",
              " 2776,\n",
              " 2782,\n",
              " 2823,\n",
              " 2833,\n",
              " 2842,\n",
              " 2843,\n",
              " 2849,\n",
              " 2860,\n",
              " 2886,\n",
              " 2892,\n",
              " 2896,\n",
              " 2902,\n",
              " 2907,\n",
              " 2931,\n",
              " 2951,\n",
              " 2963,\n",
              " 2964,\n",
              " 2982,\n",
              " 2983,\n",
              " 2989,\n",
              " 2992,\n",
              " 2998,\n",
              " 3000,\n",
              " 3003,\n",
              " 3008,\n",
              " 3040,\n",
              " 3048,\n",
              " 3076,\n",
              " 3077,\n",
              " 3080,\n",
              " 3118,\n",
              " 3124,\n",
              " 3131,\n",
              " 3148,\n",
              " 3157,\n",
              " 3160,\n",
              " 3178,\n",
              " 3197,\n",
              " 3219,\n",
              " 3220,\n",
              " 3224,\n",
              " 3225,\n",
              " 3234,\n",
              " 3236,\n",
              " 3242,\n",
              " 3249,\n",
              " 3259,\n",
              " 3264,\n",
              " 3270,\n",
              " 3272,\n",
              " 3276,\n",
              " 3286,\n",
              " 3293,\n",
              " 3305,\n",
              " 3314,\n",
              " 3320,\n",
              " 3321,\n",
              " 3325,\n",
              " 3348,\n",
              " 3353,\n",
              " 3361,\n",
              " 3365,\n",
              " 3366,\n",
              " 3375,\n",
              " 3386,\n",
              " 3389,\n",
              " 3398,\n",
              " 3412,\n",
              " 3432,\n",
              " 3442,\n",
              " 3457,\n",
              " 3461,\n",
              " 3464,\n",
              " 3474,\n",
              " 3478,\n",
              " 3481,\n",
              " 3483,\n",
              " 3493,\n",
              " 3511,\n",
              " 3514,\n",
              " 3517,\n",
              " 3518,\n",
              " 3521,\n",
              " 3528,\n",
              " 3532,\n",
              " 3535,\n",
              " 3538,\n",
              " 3579,\n",
              " 3602,\n",
              " 3613,\n",
              " 3623,\n",
              " 3628,\n",
              " 3637,\n",
              " 3640,\n",
              " 3668,\n",
              " 3671,\n",
              " 3682,\n",
              " 3685,\n",
              " 3692,\n",
              " 3713,\n",
              " 3729,\n",
              " 3735,\n",
              " 3737,\n",
              " 3745,\n",
              " 3751,\n",
              " 3764,\n",
              " 3770,\n",
              " 3782,\n",
              " 3789,\n",
              " 3804,\n",
              " 3812,\n",
              " 3823,\n",
              " 3827,\n",
              " 3830,\n",
              " 3860,\n",
              " 3862,\n",
              " 3866,\n",
              " 3870,\n",
              " 3877,\n",
              " 3897,\n",
              " 3899,\n",
              " 3911,\n",
              " 3915,\n",
              " 3917,\n",
              " 3920,\n",
              " 3935,\n",
              " 3941,\n",
              " 3967,\n",
              " 3982,\n",
              " 3988,\n",
              " 3992,\n",
              " 3995,\n",
              " 3999,\n",
              " 4020,\n",
              " 4032,\n",
              " 4038,\n",
              " 4042,\n",
              " 4051,\n",
              " 4057,\n",
              " 4065,\n",
              " 4068,\n",
              " 4080,\n",
              " 4129,\n",
              " 4130,\n",
              " 4131,\n",
              " 4132,\n",
              " 4138,\n",
              " 4139,\n",
              " 4140,\n",
              " 4159,\n",
              " 4172,\n",
              " 4173,\n",
              " 4180,\n",
              " 4189,\n",
              " 4201,\n",
              " 4208,\n",
              " 4219,\n",
              " 4239,\n",
              " 4244,\n",
              " 4245,\n",
              " 4259,\n",
              " 4260,\n",
              " 4278,\n",
              " 4282,\n",
              " 4289,\n",
              " 4308,\n",
              " 4309,\n",
              " 4318,\n",
              " 4319,\n",
              " 4322,\n",
              " 4331,\n",
              " 4360,\n",
              " 4376,\n",
              " 4377,\n",
              " 4385,\n",
              " 4394,\n",
              " 4404,\n",
              " 4410,\n",
              " 4415,\n",
              " 4421,\n",
              " 4424,\n",
              " 4426,\n",
              " 4428,\n",
              " 4441,\n",
              " 4442,\n",
              " 4444,\n",
              " 4462,\n",
              " 4463,\n",
              " 4471,\n",
              " 4477,\n",
              " 4478,\n",
              " 4481,\n",
              " 4488,\n",
              " 4489,\n",
              " 4490,\n",
              " 4502,\n",
              " 4508,\n",
              " 4517,\n",
              " 4527,\n",
              " 4535,\n",
              " 4537,\n",
              " 4546,\n",
              " 4549,\n",
              " 4555,\n",
              " 4567,\n",
              " 4571,\n",
              " 4574,\n",
              " 4575,\n",
              " 4578,\n",
              " 4579,\n",
              " 4587,\n",
              " 4595,\n",
              " 4608,\n",
              " 4616,\n",
              " 4622,\n",
              " 4624,\n",
              " 4642,\n",
              " 4647,\n",
              " 4662,\n",
              " 4673,\n",
              " 4684,\n",
              " 4694,\n",
              " 4702,\n",
              " 4704,\n",
              " 4705,\n",
              " 4706,\n",
              " 4711,\n",
              " 4714,\n",
              " 4716,\n",
              " 4719,\n",
              " 4739,\n",
              " 4741,\n",
              " 4761,\n",
              " 4762,\n",
              " 4785,\n",
              " 4794,\n",
              " 4796,\n",
              " 4809,\n",
              " 4820,\n",
              " 4823,\n",
              " 4827,\n",
              " 4830,\n",
              " 4834,\n",
              " 4843,\n",
              " 4844,\n",
              " 4859,\n",
              " 4876,\n",
              " 4880,\n",
              " 4885,\n",
              " 4888,\n",
              " 4891,\n",
              " 4893,\n",
              " 4901,\n",
              " 4903,\n",
              " 4904,\n",
              " 4920,\n",
              " 4931,\n",
              " 4947,\n",
              " 4956,\n",
              " 4963,\n",
              " 4968,\n",
              " 4970,\n",
              " 4971,\n",
              " 4978,\n",
              " 4981,\n",
              " 4984,\n",
              " 4993,\n",
              " 5005,\n",
              " 5010,\n",
              " 5011,\n",
              " 5016,\n",
              " 5018,\n",
              " 5021,\n",
              " 5028,\n",
              " 5046,\n",
              " 5073,\n",
              " 5083,\n",
              " 5085,\n",
              " 5086,\n",
              " 5088,\n",
              " 5094,\n",
              " 5095,\n",
              " 5099,\n",
              " 5111,\n",
              " 5113,\n",
              " 5115,\n",
              " 5131,\n",
              " 5139,\n",
              " 5140,\n",
              " 5142,\n",
              " 5151,\n",
              " 5152,\n",
              " 5156,\n",
              " 5165,\n",
              " 5169,\n",
              " 5172,\n",
              " 5174,\n",
              " 5180,\n",
              " 5198,\n",
              " 5210,\n",
              " 5215,\n",
              " 5219,\n",
              " 5237,\n",
              " 5244,\n",
              " 5253,\n",
              " 5256,\n",
              " 5259,\n",
              " 5260,\n",
              " 5263,\n",
              " 5277,\n",
              " 5288,\n",
              " 5294,\n",
              " 5303,\n",
              " 5312,\n",
              " 5313,\n",
              " 5324,\n",
              " 5326,\n",
              " 5335,\n",
              " 5336,\n",
              " 5339,\n",
              " 5340,\n",
              " 5344,\n",
              " 5345,\n",
              " 5355,\n",
              " 5359,\n",
              " 5360,\n",
              " 5362,\n",
              " 5368,\n",
              " 5373,\n",
              " 5376,\n",
              " 5377,\n",
              " 5383,\n",
              " 5396,\n",
              " 5424,\n",
              " 5425,\n",
              " 5430,\n",
              " 5434,\n",
              " 5435,\n",
              " 5453,\n",
              " 5459,\n",
              " 5469,\n",
              " 5471,\n",
              " 5472,\n",
              " 5482,\n",
              " 5483,\n",
              " 5500,\n",
              " 5505,\n",
              " 5508,\n",
              " 5522,\n",
              " 5554,\n",
              " 5557,\n",
              " 5559,\n",
              " 5564,\n",
              " 5574,\n",
              " 5587,\n",
              " 5589,\n",
              " 5608,\n",
              " 5612,\n",
              " 5614,\n",
              " 5615,\n",
              " 5619,\n",
              " 5620,\n",
              " 5632,\n",
              " 5638,\n",
              " 5641,\n",
              " 5643,\n",
              " 5649,\n",
              " 5667,\n",
              " 5669,\n",
              " 5678,\n",
              " 5683,\n",
              " 5684,\n",
              " 5688,\n",
              " 5689,\n",
              " 5692,\n",
              " 5699,\n",
              " 5700,\n",
              " 5701,\n",
              " 5703,\n",
              " 5715,\n",
              " 5736,\n",
              " 5740,\n",
              " 5745,\n",
              " 5755,\n",
              " 5756,\n",
              " 5757,\n",
              " 5769,\n",
              " 5782,\n",
              " 5785,\n",
              " 5809,\n",
              " 5811,\n",
              " 5823,\n",
              " 5828,\n",
              " 5830,\n",
              " 5832,\n",
              " 5862,\n",
              " 5882,\n",
              " 5883,\n",
              " 5903,\n",
              " 5906,\n",
              " 5907,\n",
              " 5913,\n",
              " 5915,\n",
              " 5916,\n",
              " 5917,\n",
              " 5933,\n",
              " 5946,\n",
              " 5947,\n",
              " 5962,\n",
              " 5967,\n",
              " 5991,\n",
              " 5994,\n",
              " 6004,\n",
              " 6010,\n",
              " 6016,\n",
              " 6026,\n",
              " 6031,\n",
              " 6041,\n",
              " 6042,\n",
              " 6051,\n",
              " 6053,\n",
              " 6057,\n",
              " 6066,\n",
              " 6068,\n",
              " 6075,\n",
              " 6101,\n",
              " 6107,\n",
              " 6120,\n",
              " 6140,\n",
              " 6146,\n",
              " 6148,\n",
              " 6151,\n",
              " 6155,\n",
              " 6160,\n",
              " 6178,\n",
              " 6182,\n",
              " 6190,\n",
              " 6197,\n",
              " 6200,\n",
              " 6216,\n",
              " 6225,\n",
              " 6229,\n",
              " 6230,\n",
              " 6233,\n",
              " 6241,\n",
              " 6247,\n",
              " 6253,\n",
              " 6262,\n",
              " 6263,\n",
              " 6268,\n",
              " 6270,\n",
              " 6287,\n",
              " 6293,\n",
              " 6295,\n",
              " 6318,\n",
              " 6327,\n",
              " 6332,\n",
              " 6334,\n",
              " 6336,\n",
              " 6338,\n",
              " 6339,\n",
              " 6352,\n",
              " 6355,\n",
              " 6357,\n",
              " 6358,\n",
              " 6363,\n",
              " 6364,\n",
              " 6379,\n",
              " 6380,\n",
              " 6406,\n",
              " 6407,\n",
              " 6409,\n",
              " 6414,\n",
              " 6421,\n",
              " 6422,\n",
              " 6424,\n",
              " 6426,\n",
              " 6428,\n",
              " 6432,\n",
              " 6447,\n",
              " 6451,\n",
              " 6464,\n",
              " 6465,\n",
              " 6481,\n",
              " 6488,\n",
              " 6489,\n",
              " 6491,\n",
              " 6512,\n",
              " 6517,\n",
              " 6518,\n",
              " 6520,\n",
              " 6522,\n",
              " 6531,\n",
              " 6539,\n",
              " 6541,\n",
              " 6560,\n",
              " 6562,\n",
              " 6572,\n",
              " 6578,\n",
              " 6586,\n",
              " 6590,\n",
              " 6595,\n",
              " 6599,\n",
              " 6602,\n",
              " 6603,\n",
              " 6627,\n",
              " 6631,\n",
              " 6632,\n",
              " 6640,\n",
              " 6647,\n",
              " 6651,\n",
              " 6664,\n",
              " 6675,\n",
              " 6692,\n",
              " 6709,\n",
              " 6710,\n",
              " 6715,\n",
              " 6721,\n",
              " 6725,\n",
              " 6730,\n",
              " 6733,\n",
              " 6744,\n",
              " 6747,\n",
              " 6749,\n",
              " 6753,\n",
              " 6760,\n",
              " 6764,\n",
              " 6765,\n",
              " 6773,\n",
              " 6777,\n",
              " 6780,\n",
              " 6790,\n",
              " 6792,\n",
              " 6800,\n",
              " 6809,\n",
              " 6811,\n",
              " 6819,\n",
              " 6824,\n",
              " 6842,\n",
              " 6846,\n",
              " 6860,\n",
              " 6862,\n",
              " 6873,\n",
              " 6901,\n",
              " 6914,\n",
              " 6920,\n",
              " 6935,\n",
              " 6936,\n",
              " 6941,\n",
              " 6943,\n",
              " 6945,\n",
              " 6957,\n",
              " 6964,\n",
              " 6973,\n",
              " 6981,\n",
              " 6990,\n",
              " 6996,\n",
              " 6998,\n",
              " 7022,\n",
              " 7028,\n",
              " 7035,\n",
              " 7040,\n",
              " 7048,\n",
              " 7049,\n",
              " 7069,\n",
              " 7090,\n",
              " 7095,\n",
              " 7103,\n",
              " 7104,\n",
              " 7116,\n",
              " 7123,\n",
              " 7124,\n",
              " 7129,\n",
              " 7139,\n",
              " 7143,\n",
              " 7145,\n",
              " 7150,\n",
              " 7159,\n",
              " 7167,\n",
              " 7174,\n",
              " 7179,\n",
              " 7201,\n",
              " 7205,\n",
              " 7220,\n",
              " 7221,\n",
              " 7224,\n",
              " 7228,\n",
              " 7232,\n",
              " 7239,\n",
              " 7247,\n",
              " 7251,\n",
              " 7275,\n",
              " 7277,\n",
              " 7307,\n",
              " 7318,\n",
              " 7319,\n",
              " 7357,\n",
              " 7361,\n",
              " 7367,\n",
              " 7393,\n",
              " 7396,\n",
              " 7420,\n",
              " 7424,\n",
              " 7452,\n",
              " 7455,\n",
              " 7476,\n",
              " 7489,\n",
              " 7498,\n",
              " 7500,\n",
              " 7503,\n",
              " 7504,\n",
              " 7510,\n",
              " ...]"
            ]
          },
          "metadata": {},
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#train_imgs_features.values()"
      ],
      "metadata": {
        "id": "X9HqXmxZk-sB"
      },
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Define the paths to your input files\n",
        "question_file_path = \"/content/v2_OpenEnded_mscoco_train2014_questions.json\"\n",
        "annotation_file_path = \"/content/v2_mscoco_train2014_annotations.json\"\n",
        "\n",
        "# Load the question file and extract the question data\n",
        "with open(question_file_path, \"r\") as f:\n",
        "    question_data = json.load(f)[\"questions\"]\n",
        "\n",
        "# Load the annotation file and extract the ground truth answers\n",
        "with open(annotation_file_path, \"r\") as f:\n",
        "    annotation_data = json.load(f)[\"annotations\"]"
      ],
      "metadata": {
        "id": "rJfeRCBUeqWo"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Group the question and answer data by img_id using a dictionary comprehension\n",
        "qa_data = {a['image_id']: {'questions': [], 'answers': []} for a in annotation_data}"
      ],
      "metadata": {
        "id": "5LY2Up45-Tcz"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "for q in question_data:\n",
        "    qa_data[q['image_id']]['questions'].append(q['question'])"
      ],
      "metadata": {
        "id": "ZRrIqSSY-TKz"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "for a in annotation_data:\n",
        "    qa_data[a['image_id']]['answers'].append(a['multiple_choice_answer'])"
      ],
      "metadata": {
        "id": "zELfgmxv-TE3"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Generate the input data using a list comprehension\n",
        "input_data = [{\"image\": train_imgs_features[key], \"question\": q, \"answer\": a} \n",
        "              for key in list(train_imgs_features.keys()) \n",
        "              for q, a in zip(qa_data[key]['questions'], qa_data[key]['answers'])]"
      ],
      "metadata": {
        "id": "Lniwcq6a-S_J"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(input_data[150])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ooa_uX0k-SxZ",
        "outputId": "a47a37c7-eaec-4003-bb14-a5fb8f118243"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "{'image': array([0.44045335, 0.50492954, 0.14766406, ..., 0.22389303, 0.21844536,\n",
            "       0.28869957], dtype=float32), 'question': 'Where was picture taken?', 'answer': 'park'}\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Load the pre-trained models\n",
        "efnet_model = EfficientNetB2(weights='imagenet')\n",
        "bert_tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')\n",
        "bert_model = TFBertForQuestionAnswering.from_pretrained('bert-base-uncased')"
      ],
      "metadata": {
        "id": "JyA2-dSley7I",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 269,
          "referenced_widgets": [
            "5fb3ce547c094be5a95ab91dec35239b",
            "fac5192766354c6ea714135820ba9baa",
            "1ee627fc31eb4038a96e8e6e74319fd8",
            "f6c46660547e4b89aaee1771b8dfcd4e",
            "2c6f50e07320456fadf95fcfffe82fe5",
            "0ed4b7683c8f4148bdda6c9d8de0105f",
            "6a97d46b971240fc89710ced32b4fa0f",
            "19f403239d8c416992853077fb256169",
            "0bae9c1192004902ab3b6405ec0100c0",
            "3b25874a1eb34e69b1a254ee3cd0dcaa",
            "6d0260d6a39f479ea303252f5d4f5fac",
            "89dc8cdaff614483b8aa6bb9cffc1b37",
            "94cd1383bbdf4be9beb5dc77313f88b2",
            "ca69bb68b6144a60b3aa5345c564bced",
            "737ac1589afd4dbbb38b5290b6e1a79f",
            "4bd737b8c61e4f3087e03ed1a991b9d2",
            "9e7a3922a041437f9015174fe29fefcc",
            "3882812a3169426886345f87b56e389c",
            "3c8ecd5fed1044aa9ef96c47898d0f25",
            "13112cff5cbb4c44a0af17cab75009c1",
            "c211d7eda87e490db0588618def13950",
            "d398c47e9a224f2b8d4027ba428feadf",
            "1092e1708f4c4ac2bc69a8bff3112350",
            "6e816e1c53684871bf0146478950995d",
            "c24e60db68af4836b9832e838f6e7e29",
            "95ff807dfc7e47c5a5afc7e94140d0ce",
            "4c563d3c07dc4406a7bbcac048f659ac",
            "ce687fe31c9543b4b981f20f89e4b172",
            "85780fdfeeb04ffbb080a2f1c8437253",
            "441f326476344d9da8475b50af4feeea",
            "9e115f7eff0b4f6fb1fa0a1238c80c1f",
            "210bc13fa03e4abe8a3c6cf4b272fec6",
            "70d7063610f64c96be56dc4446f164e3",
            "fd2c68dcf04746758952d4a83800db0b",
            "21b7113e6454424591ca81e4e6b0d849",
            "a301a339beb5479cbde2c3659723e5a8",
            "bde10fda39344fe2bd5fe57030acb2f7",
            "7c855d452e554e83aa2d83a3fdb0cce6",
            "bc1c306f2ca74d5aa7ab5dd894ef3a15",
            "71ab4f0156e343c286fa125c24a9bfbc",
            "58c9dbd342eb4183bfce38c8007a8f57",
            "7569e61018144ff4a5287e6c3c5f24b3",
            "8e02377b4627418791bc440762a5dd90",
            "0ad8638e33054feb88f120bf23be59e6"
          ]
        },
        "outputId": "ad947d1f-6eb4-4a75-894d-ca1e583c7383"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading data from https://github.com/Callidior/keras-applications/releases/download/efficientnet/efficientnet-b2_weights_tf_dim_ordering_tf_kernels_autoaugment.h5\n",
            "37424016/37424016 [==============================] - 0s 0us/step\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading (…)solve/main/vocab.txt:   0%|          | 0.00/232k [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "5fb3ce547c094be5a95ab91dec35239b"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading (…)okenizer_config.json:   0%|          | 0.00/28.0 [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "89dc8cdaff614483b8aa6bb9cffc1b37"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading (…)lve/main/config.json:   0%|          | 0.00/570 [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "1092e1708f4c4ac2bc69a8bff3112350"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading tf_model.h5:   0%|          | 0.00/536M [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "fd2c68dcf04746758952d4a83800db0b"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "All model checkpoint layers were used when initializing TFBertForQuestionAnswering.\n",
            "\n",
            "Some layers of TFBertForQuestionAnswering were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['qa_outputs']\n",
            "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Tokenize and encode the question data using the appropriate tokenizer\n",
        "encoded_data = []\n",
        "for d in input_data:\n",
        "    question_tokens = bert_tokenizer(d[\"question\"], padding=True, truncation=True, return_tensors=\"tf\")\n",
        "    encoded_data.append({\"image\": d[\"image\"], \"question\": question_tokens, \"answer\": d[\"answer\"]})\n",
        "\n"
      ],
      "metadata": {
        "id": "IF6jHXhAHoYs"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Pass the preprocessed input data to the pre-trained models to get the predicted answers\n",
        "bert_outputs = []\n",
        "for d in encoded_data:\n",
        "    #efnet_output = tf.keras.layers.Flatten()(d[\"image\"])\n",
        "    efnet_output = d[\"image\"]\n",
        "    print(efnet_output)\n",
        "    bert_input = {\"input_ids\": d[\"question\"][\"input_ids\"], \"attention_mask\": d[\"question\"][\"attention_mask\"]}#, 'image': efnet_output }\n",
        "    print(bert_input)\n",
        "    bert_output = bert_model(bert_input)\n",
        "    bert_outputs.append(tf.argmax(bert_output.start_logits, axis=-1).numpy())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-suuzOXjP5h3",
        "outputId": "0941ca83-bbb6-4db4-9523-844eaf34ee0c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1;30;43mStreaming output truncated to the last 5000 lines.\u001b[0m\n",
            "array([[  101,  2054,  2003,  2369,  1996, 12586,  2121,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.36214322 0.11902103 0.22257867 ... 0.35679483 0.26675975 0.2538435 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2045,  1037,  5404, 14757,  2006,  1996,  2795,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.36214322 0.11902103 0.22257867 ... 0.35679483 0.26675975 0.2538435 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996, 18097,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.12593982 0.40667284 0.49511993 ... 0.37841183 0.1585562  0.11781067]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2073, 2024, 1996, 6007, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.12593982 0.40667284 0.49511993 ... 0.37841183 0.1585562  0.11781067]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  2665,  6910,  2006,  1996, 10005,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.12593982 0.40667284 0.49511993 ... 0.37841183 0.1585562  0.11781067]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 2793, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.12593982 0.40667284 0.49511993 ... 0.37841183 0.1585562  0.11781067]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 2023, 2793, 2081, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.12593982 0.40667284 0.49511993 ... 0.37841183 0.1585562  0.11781067]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[ 101, 2054, 2003, 2330, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2928254  0.08122127 0.18667471 ... 0.14207599 0.12555483 0.40733457]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 1037, 3584, 2341, 2005, 1996, 6457, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2928254  0.08122127 0.18667471 ... 0.14207599 0.12555483 0.40733457]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 1037, 5259, 2006, 1996, 2813, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2928254  0.08122127 0.18667471 ... 0.14207599 0.12555483 0.40733457]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 2715, 5723, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2928254  0.08122127 0.18667471 ... 0.14207599 0.12555483 0.40733457]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996, 11848,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.4165188  0.6525128  0.25541773 ... 0.40843898 0.02724206 0.2362607 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 2105, 2010, 3300, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.4165188  0.6525128  0.25541773 ... 0.40843898 0.02724206 0.2362607 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 1996, 2158, 3173, 2242, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.4165188  0.6525128  0.25541773 ... 0.40843898 0.02724206 0.2362607 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 3087, 1999, 2023, 6302, 4147, 7877, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08546525 0.08798628 0.07114138 ... 0.14384767 0.5072422  0.08542916]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 3609, 1997, 1996, 2300, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08546525 0.08798628 0.07114138 ... 0.14384767 0.5072422  0.08542916]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2106,  2002,  4558,  2010, 14175,  6277,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08546525 0.08798628 0.07114138 ... 0.14384767 0.5072422  0.08542916]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996,  2158,  4147,  1037,  4954, 28880,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2589046  0.23102382 0.12278084 ... 0.6818261  0.1980409  0.63807297]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2071,  2122,  2022, 24881,  1998, 16130,  1011, 13017,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2589046  0.23102382 0.12278084 ... 0.6818261  0.1980409  0.63807297]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 3120, 1997, 5250, 1999, 2023, 7954, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2589046  0.23102382 0.12278084 ... 0.6818261  0.1980409  0.63807297]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 3482, 2440, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2589046  0.23102382 0.12278084 ... 0.6818261  0.1980409  0.63807297]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023,  3262,  4086,  2030,  7842, 14550,  2100,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2589046  0.23102382 0.12278084 ... 0.6818261  0.1980409  0.63807297]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116,  7967,  2123, 16446,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2589046  0.23102382 0.12278084 ... 0.6818261  0.1980409  0.63807297]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 15), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  6240,  2003,  2006,  2327,  1997,  1996,\n",
            "         1017, 11035,  2123, 16446,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 15), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.61019    0.10246309 0.30693337 ... 0.13341403 0.35431325 0.5315971 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2008,  1037, 24471, 13290,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.61019    0.10246309 0.30693337 ... 0.13341403 0.35431325 0.5315971 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023, 28249,  2005,  1037,  3287,  2030,  2931,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.61019    0.10246309 0.30693337 ... 0.13341403 0.35431325 0.5315971 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 2023, 2341, 2441, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.15375045 0.13837427 0.31090012 ... 0.10193525 0.17972396 0.69707936]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2024,  2122, 21674,  4198,  2000, 22365,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.15375045 0.13837427 0.31090012 ... 0.10193525 0.17972396 0.69707936]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2024,  1996, 21674,  2035,  1996,  2168,  3609,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.15375045 0.13837427 0.31090012 ... 0.10193525 0.17972396 0.69707936]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2024, 21674,  1996,  2069,  5167,  8250,  1999,  1996,\n",
            "         2282,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.12686396 0.47143224 0.51089525 ... 0.16804446 0.95040554 0.71489096]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 5472, 5744, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.12686396 0.47143224 0.51089525 ... 0.16804446 0.95040554 0.71489096]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 2023, 2158, 3394, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.12686396 0.47143224 0.51089525 ... 0.16804446 0.95040554 0.71489096]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 2158, 2725, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20749743 0.4958287  0.12962943 ... 0.3128225  0.05302812 1.4727956 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2024,  1996, 18108,  2665,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20749743 0.4958287  0.12962943 ... 0.3128225  0.05302812 1.4727956 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2828,  1997, 27619,  2024,  2122,  1006,  3384,\n",
            "         2030,  3536,  1007,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20749743 0.4958287  0.12962943 ... 0.3128225  0.05302812 1.4727956 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 18108,  2024,  2045,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04715315 0.11357324 0.2996875  ... 0.05887755 0.08636704 0.1688301 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 1996, 2812, 4147, 5337, 4929, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04715315 0.11357324 0.2996875  ... 0.05887755 0.08636704 0.1688301 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2723, 2081, 1997, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04715315 0.11357324 0.2996875  ... 0.05887755 0.08636704 0.1688301 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023,  1037, 16125, 22861,  4160,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04715315 0.11357324 0.2996875  ... 0.05887755 0.08636704 0.1688301 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 2813, 2369, 1996, 2273, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5541048  1.3662796  0.90109277 ... 0.18756413 0.5457561  0.15831971]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2828, 1997, 9476, 2839, 2003, 2008, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5541048  1.3662796  0.90109277 ... 0.18756413 0.5457561  0.15831971]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023,  2833,  4086,  2030, 14768,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5541048  1.3662796  0.90109277 ... 0.18756413 0.5457561  0.15831971]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054, 21183,  6132,  4014,  2003,  1996,  2839,  3173,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.36155447 0.6558676  0.04408472 ... 0.16879055 0.07858763 0.3053527 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 3609, 1997, 1996, 3797, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.36155447 0.6558676  0.04408472 ... 0.16879055 0.07858763 0.3053527 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 3609, 1997, 1996, 2158, 1005, 1055, 3797,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.36155447 0.6558676  0.04408472 ... 0.16879055 0.07858763 0.3053527 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996,  2158,  4550, 20665,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.36155447 0.6558676  0.04408472 ... 0.16879055 0.07858763 0.3053527 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2323, 1996, 2158, 4929, 2008, 5495, 2007, 2008, 3797, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.36155447 0.6558676  0.04408472 ... 0.16879055 0.07858763 0.3053527 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 2122, 4253, 6413, 2005, 2147, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.63284844 0.25932294 0.08170165 ... 0.05100683 0.3325756  0.7231585 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023,  1037, 14380,  9648,  3233,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.63284844 0.25932294 0.08170165 ... 0.05100683 0.3325756  0.7231585 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2023, 1037, 4825, 2030, 2797, 3829, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.63284844 0.25932294 0.08170165 ... 0.05100683 0.3325756  0.7231585 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2828, 1997, 4497, 2003, 2023, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.41647172 0.14482802 0.1142258  ... 0.14358073 0.25374457 0.6066531 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 6468, 2024, 2045, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.41647172 0.14482802 0.1142258  ... 0.14358073 0.25374457 0.6066531 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 1996, 6468, 6530, 2030, 4550, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.41647172 0.14482802 0.1142258  ... 0.14358073 0.25374457 0.6066531 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2064, 2017, 4521, 2023, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.41647172 0.14482802 0.1142258  ... 0.14358073 0.25374457 0.6066531 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2024, 2122, 2613, 6468, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11684011 0.02149295 0.21564737 ... 0.17184085 0.2010316  1.0278776 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 1037, 4037, 4769, 2556, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11684011 0.02149295 0.21564737 ... 0.17184085 0.2010316  1.0278776 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  1996, 25791,  2193,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11684011 0.02149295 0.21564737 ... 0.17184085 0.2010316  1.0278776 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996, 14878,  2183,  3435,  2030,  4030,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18991838 0.6771289  0.05563262 ... 0.3161617  0.2944208  1.0232544 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2515,  1996,  8638,  2031, 25007,  7318,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18991838 0.6771289  0.05563262 ... 0.3161617  0.2944208  1.0232544 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2073,  2003,  1996, 11190,  5307,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18991838 0.6771289  0.05563262 ... 0.3161617  0.2944208  1.0232544 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 8638, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18991838 0.6771289  0.05563262 ... 0.3161617  0.2944208  1.0232544 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2502,  2003,  1996, 11190,  1005,  1055,  4451,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6209857  0.49869704 0.16057248 ... 0.11442366 0.21381837 0.00262007]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 4338, 1997, 1996, 3696, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6209857  0.49869704 0.16057248 ... 0.11442366 0.21381837 0.00262007]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 1996, 3696, 1999, 2394, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6209857  0.49869704 0.16057248 ... 0.11442366 0.21381837 0.00262007]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 2023, 1996, 3915, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6209857  0.49869704 0.16057248 ... 0.11442366 0.21381837 0.00262007]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2171, 2006, 1996, 3696, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6209857  0.49869704 0.16057248 ... 0.11442366 0.21381837 0.00262007]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2171, 1997, 1996, 2395, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6209857  0.49869704 0.16057248 ... 0.11442366 0.21381837 0.00262007]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 2644, 3696, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6209857  0.49869704 0.16057248 ... 0.11442366 0.21381837 0.00262007]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 1996, 3696, 2758, 2644, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6209857  0.49869704 0.16057248 ... 0.11442366 0.21381837 0.00262007]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 4338, 2003, 2023, 3696, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6209857  0.49869704 0.16057248 ... 0.11442366 0.21381837 0.00262007]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 2395, 5751, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6209857  0.49869704 0.16057248 ... 0.11442366 0.21381837 0.00262007]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 3696, 4987, 2205, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6209857  0.49869704 0.16057248 ... 0.11442366 0.21381837 0.00262007]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 6087, 2024, 1996, 3696, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6209857  0.49869704 0.16057248 ... 0.11442366 0.21381837 0.00262007]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 2395, 3696, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6209857  0.49869704 0.16057248 ... 0.11442366 0.21381837 0.00262007]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 4111, 4871, 2024, 2556, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6209857  0.49869704 0.16057248 ... 0.11442366 0.21381837 0.00262007]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 2369, 1996, 3696, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6209857  0.49869704 0.16057248 ... 0.11442366 0.21381837 0.00262007]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2515, 1996, 3696, 2360, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.39300483 0.0740466  0.2377676  ... 0.07797575 0.3268307  0.00121963]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  1996,  3609,  1997,  1996, 23801,  1005,\n",
            "         1055, 10412,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.39300483 0.0740466  0.2377676  ... 0.07797575 0.3268307  0.00121963]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2998, 2276, 2003, 3491, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.39300483 0.0740466  0.2377676  ... 0.07797575 0.3268307  0.00121963]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2003,  1996,  2158, 11820,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.39300483 0.0740466  0.2377676  ... 0.07797575 0.3268307  0.00121963]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2023, 1037, 2350, 2223, 2208, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.39300483 0.0740466  0.2377676  ... 0.07797575 0.3268307  0.00121963]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2024, 1996, 2136, 6087, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.4265048  0.3082495  1.0461245  ... 0.24339658 0.21568573 0.11724532]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2515, 1996, 3392, 2031, 5909, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.4265048  0.3082495  1.0461245  ... 0.24339658 0.21568573 0.11724532]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2364, 3609, 2003, 1996, 7997, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.4265048  0.3082495  1.0461245  ... 0.24339658 0.21568573 0.11724532]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 2160, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.4265048  0.3082495  1.0461245  ... 0.24339658 0.21568573 0.11724532]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2003,  1996,  2879, 12447,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.56363773 0.24743348 0.80460703 ... 0.20469221 0.868385   0.4010834 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 2023, 2450, 6023, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.56363773 0.24743348 0.80460703 ... 0.20469221 0.868385   0.4010834 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2079,  2027,  2562, 10733, 12760,  2006,  2192,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.56363773 0.24743348 0.80460703 ... 0.20469221 0.868385   0.4010834 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2001, 2023, 3861, 2579, 2648, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.56363773 0.24743348 0.80460703 ... 0.20469221 0.868385   0.4010834 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 3609, 1997, 1996, 2450, 1005, 1055, 3797,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.56363773 0.24743348 0.80460703 ... 0.20469221 0.868385   0.4010834 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2515, 1996, 3203, 3711, 2000, 2022, 8434, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.56363773 0.24743348 0.80460703 ... 0.20469221 0.868385   0.4010834 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 2111, 2024, 5710, 1999, 2023, 3861, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.56363773 0.24743348 0.80460703 ... 0.20469221 0.868385   0.4010834 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 1996, 2450, 2635, 3964, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.56363773 0.24743348 0.80460703 ... 0.20469221 0.868385   0.4010834 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 3203, 4527, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.56363773 0.24743348 0.80460703 ... 0.20469221 0.868385   0.4010834 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2024, 1996, 4624, 2015, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.56363773 0.24743348 0.80460703 ... 0.20469221 0.868385   0.4010834 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2006, 1996, 2450, 1005, 1055, 2132, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.56363773 0.24743348 0.80460703 ... 0.20469221 0.868385   0.4010834 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116,  2024,  4147, 16717,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.56363773 0.24743348 0.80460703 ... 0.20469221 0.868385   0.4010834 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2214, 2003, 2023, 2450, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.56363773 0.24743348 0.80460703 ... 0.20469221 0.868385   0.4010834 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2040, 2003, 4147, 1037, 6178, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.56363773 0.24743348 0.80460703 ... 0.20469221 0.868385   0.4010834 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023,  2711,  1037, 10563,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.56363773 0.24743348 0.80460703 ... 0.20469221 0.868385   0.4010834 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 2014, 2606, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.8814841e-01 3.1938635e-02 8.7709621e-02 ... 5.0324900e-04 6.1657774e-01\n",
            " 4.3716499e-01]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 4111, 2003, 2023, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.8814841e-01 3.1938635e-02 8.7709621e-02 ... 5.0324900e-04 6.1657774e-01\n",
            " 4.3716499e-01]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023,  4111,  1037, 25476,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.8814841e-01 3.1938635e-02 8.7709621e-02 ... 5.0324900e-04 6.1657774e-01\n",
            " 4.3716499e-01]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2023, 4111, 1999, 2049, 3128, 6552, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.53808296 0.25691515 0.1381445  ... 0.06038668 0.07816332 0.15255734]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 2045, 1037, 2958, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.53808296 0.25691515 0.1381445  ... 0.06038668 0.07816332 0.15255734]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2029,  2482,  2006,  1996,  3345,  2003, 14962,  1037,\n",
            "         2732,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.53808296 0.25691515 0.1381445  ... 0.06038668 0.07816332 0.15255734]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 5568, 4206, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.53808296 0.25691515 0.1381445  ... 0.06038668 0.07816332 0.15255734]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 3345, 3765, 2024, 2045, 1999, 2023, 3496, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.53808296 0.25691515 0.1381445  ... 0.06038668 0.07816332 0.15255734]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 2440, 3345, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.53808296 0.25691515 0.1381445  ... 0.06038668 0.07816332 0.15255734]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 3765, 2191, 2039, 2023, 3345, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.53808296 0.25691515 0.1381445  ... 0.06038668 0.07816332 0.15255734]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 3609, 1997, 1996, 3345, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.53808296 0.25691515 0.1381445  ... 0.06038668 0.07816332 0.15255734]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 3609, 1997, 1996, 2877, 3345, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.53808296 0.25691515 0.1381445  ... 0.06038668 0.07816332 0.15255734]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 3765, 2515, 1996, 3345, 2031, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.53808296 0.25691515 0.1381445  ... 0.06038668 0.07816332 0.15255734]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2515, 2023, 3345, 2031, 1037, 9298, 9541, 3366, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.53808296 0.25691515 0.1381445  ... 0.06038668 0.07816332 0.15255734]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 6636, 3345, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20933427 0.07976104 0.13651842 ... 0.         0.10918948 0.2426811 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2073, 2003, 1996, 6097, 1998, 8808, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20933427 0.07976104 0.13651842 ... 0.         0.10918948 0.2426811 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2073, 2001, 2023, 6302, 2579, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20933427 0.07976104 0.13651842 ... 0.         0.10918948 0.2426811 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2024,  1996,  2123, 16446,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20933427 0.07976104 0.13651842 ... 0.         0.10918948 0.2426811 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 2327, 2123, 4904, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20933427 0.07976104 0.13651842 ... 0.         0.10918948 0.2426811 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2515, 2023, 3573, 2569, 4697, 1999, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20933427 0.07976104 0.13651842 ... 0.         0.10918948 0.2426811 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  1996,  2317,  2813, 12183,  3065,  2054,  7957,  1997,\n",
            "         2833,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20933427 0.07976104 0.13651842 ... 0.         0.10918948 0.2426811 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2171, 1997, 2023, 4825, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20933427 0.07976104 0.13651842 ... 0.         0.10918948 0.2426811 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2023, 1037, 6302, 1999, 1037, 3573, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20933427 0.07976104 0.13651842 ... 0.         0.10918948 0.2426811 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 1037, 7026, 2006, 1996, 2813, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20933427 0.07976104 0.13651842 ... 0.         0.10918948 0.2426811 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2171, 1997, 1996, 4825, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20933427 0.07976104 0.13651842 ... 0.         0.10918948 0.2426811 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2097, 2027, 5271, 2833, 2045, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20933427 0.07976104 0.13651842 ... 0.         0.10918948 0.2426811 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2833, 2003, 1999, 1996, 2157, 3953, 3420, 1997, 1996,\n",
            "        3746, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20933427 0.07976104 0.13651842 ... 0.         0.10918948 0.2426811 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116,  7957,  1997, 15890,  2015,  2024,  2853,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2204485  0.08797251 0.351371   ... 0.30863953 0.9847307  0.2677945 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2024,  2119,  2273,  1999,  2023,  6302,  4147, 16717,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2204485  0.08797251 0.351371   ... 0.30863953 0.9847307  0.2677945 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2024, 2027, 3564, 2006, 5194, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2204485  0.08797251 0.351371   ... 0.30863953 0.9847307  0.2677945 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=\n",
            "array([[ 101, 2029, 3586, 2071, 2022, 1000, 4830, 9397, 2571, 1011, 3897,\n",
            "        1000, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2204485  0.08797251 0.351371   ... 0.30863953 0.9847307  0.2677945 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[  101,  2106,  2028,  1997,  1996,  5194, 13433,  2080,  2006,\n",
            "         1996,  2598,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2204485  0.08797251 0.351371   ... 0.30863953 0.9847307  0.2677945 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2024, 2122, 2273, 1005, 1055, 9518, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.46437162 0.5422312  0.20665824 ... 1.6655042  0.12038884 0.14596476]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2023, 1996, 2203, 1997, 1996, 2650, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.46437162 0.5422312  0.20665824 ... 1.6655042  0.12038884 0.14596476]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2001, 2023, 3861, 2579, 1999, 2621, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.46437162 0.5422312  0.20665824 ... 1.6655042  0.12038884 0.14596476]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 3345, 3048, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.46437162 0.5422312  0.20665824 ... 1.6655042  0.12038884 0.14596476]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 3345, 3228, 2125, 5492, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.46437162 0.5422312  0.20665824 ... 1.6655042  0.12038884 0.14596476]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 3696, 2079, 2017, 2156, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1596154  0.91334605 0.46856797 ... 0.44189322 0.614275   0.8634095 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 2182, 4377, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1596154  0.91334605 0.46856797 ... 0.44189322 0.614275   0.8634095 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 2450, 3173, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1596154  0.91334605 0.46856797 ... 0.44189322 0.614275   0.8634095 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 4562, 4142, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.24807267 0.53428435 0.14926736 ... 0.5418551  0.4888813  0.2811731 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023,  1037, 17260,  2380,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.24807267 0.53428435 0.14926736 ... 0.5418551  0.4888813  0.2811731 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 6007, 2024, 2216, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.24807267 0.53428435 0.14926736 ... 0.5418551  0.4888813  0.2811731 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2129,  2116,  2024, 10080,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3814867  0.21678987 0.18804795 ... 0.554417   0.02444535 0.16664046]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2835, 2006, 1996, 3242, 2081, 1997, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3814867  0.21678987 0.18804795 ... 0.554417   0.02444535 0.16664046]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2006, 1996, 2849, 1997, 1996, 3242, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3814867  0.21678987 0.18804795 ... 0.554417   0.02444535 0.16664046]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2051, 2003, 2009, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5698416  0.08030678 0.7361839  ... 0.75475466 0.22946496 0.44690162]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 3293, 4683, 3039, 2006, 2023, 2395, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5698416  0.08030678 0.7361839  ... 0.75475466 0.22946496 0.44690162]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2003,  2045,  1037, 24581,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5698416  0.08030678 0.7361839  ... 0.75475466 0.22946496 0.44690162]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 3712, 2630, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5698416  0.08030678 0.7361839  ... 0.75475466 0.22946496 0.44690162]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996,  3392, 12283,  2084,  1996,  2311,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5698416  0.08030678 0.7361839  ... 0.75475466 0.22946496 0.44690162]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2515, 1996, 2193, 1019, 2812, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5698416  0.08030678 0.7361839  ... 0.75475466 0.22946496 0.44690162]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 2045, 8044, 1999, 1996, 6302, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5698416  0.08030678 0.7361839  ... 0.75475466 0.22946496 0.44690162]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2828,  1997,  2346,  2003, 17099,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.4972101  0.29130214 0.31313533 ... 0.2081664  0.09673253 0.43250507]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 2111, 2024, 3797, 3238, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.4972101  0.29130214 0.31313533 ... 0.2081664  0.09673253 0.43250507]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2024,  1996,  2111, 20120,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.4972101  0.29130214 0.31313533 ... 0.2081664  0.09673253 0.43250507]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2116, 17260, 15271,  2024,  2045,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.4972101  0.29130214 0.31313533 ... 0.2081664  0.09673253 0.43250507]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609, 11344,  2024,  2027,  2035,  4147,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3635245  0.05146465 0.04199702 ... 0.01462426 0.6397097  1.1125746 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2003, 2023, 8875, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3635245  0.05146465 0.04199702 ... 0.01462426 0.6397097  1.1125746 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 2604, 2157, 2217, 2039, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3635245  0.05146465 0.04199702 ... 0.01462426 0.6397097  1.1125746 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 7787, 2079, 2017, 2156, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.47900504 0.04637977 1.3560358  ... 0.2731806  0.81803775 0.78617936]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2064, 2017, 3191, 2122, 2808, 2006, 2785, 2571, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.47900504 0.04637977 1.3560358  ... 0.2731806  0.81803775 0.78617936]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116,  2808,  2024, 16934,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.47900504 0.04637977 1.3560358  ... 0.2731806  0.81803775 0.78617936]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2006, 2327, 1997, 1996, 2808, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.05644447 0.77203906 0.6512862  ... 0.41282812 0.492337   0.33823743]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2024, 2027, 2383, 4157, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.05644447 0.77203906 0.6512862  ... 0.41282812 0.492337   0.33823743]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2024,  2045,  2151, 16324,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.05644447 0.77203906 0.6512862  ... 0.41282812 0.492337   0.33823743]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2833, 2003, 2006, 1996, 3953, 5127, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5466646  0.16714603 0.16712934 ... 0.         0.7041007  0.05818811]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2023, 2711, 2667, 2000, 2202, 1037, 3861, 1997,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5466646  0.16714603 0.16712934 ... 0.         0.7041007  0.05818811]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 2711, 2503, 2030, 2648, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5466646  0.16714603 0.16712934 ... 0.         0.7041007  0.05818811]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2023, 1996, 2190, 2126, 2000, 2202, 1037, 2969, 2666,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.24987954 0.23497903 0.06919612 ... 0.0529801  1.5093317  0.59288764]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2111, 2024, 1999, 1996, 3746, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.24987954 0.23497903 0.06919612 ... 0.0529801  1.5093317  0.59288764]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2515, 2023, 2298, 2066, 2019, 3199, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.24987954 0.23497903 0.06919612 ... 0.0529801  1.5093317  0.59288764]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2023, 1037, 3773, 3239, 3899, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.32915854 0.08139846 0.41436577 ... 0.17102067 0.09254717 0.1166945 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 4744, 2109, 2411, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.32915854 0.08139846 0.41436577 ... 0.17102067 0.09254717 0.1166945 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2785,  1997,  4744,  2003, 15885,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.32915854 0.08139846 0.41436577 ... 0.17102067 0.09254717 0.1166945 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[ 101, 2054, 2003, 2023, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.32915854 0.08139846 0.41436577 ... 0.17102067 0.09254717 0.1166945 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 13310,  2024,  2006,  1996,  2665,  4744,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.32915854 0.08139846 0.41436577 ... 0.17102067 0.09254717 0.1166945 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 4744, 2303, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.32915854 0.08139846 0.41436577 ... 0.17102067 0.09254717 0.1166945 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 8154, 2006, 1996, 2217, 1997, 1996, 4744,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.32915854 0.08139846 0.41436577 ... 0.17102067 0.09254717 0.1166945 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996, 21519,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.32915854 0.08139846 0.41436577 ... 0.17102067 0.09254717 0.1166945 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2515, 1996, 4744, 2298, 4704, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.32915854 0.08139846 0.41436577 ... 0.17102067 0.09254717 0.1166945 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 13310,  2064,  2017,  2156,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.32915854 0.08139846 0.41436577 ... 0.17102067 0.09254717 0.1166945 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 4170, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.32915854 0.08139846 0.41436577 ... 0.17102067 0.09254717 0.1166945 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 4597, 2024, 2006, 1996, 4744, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.32915854 0.08139846 0.41436577 ... 0.17102067 0.09254717 0.1166945 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 7787, 2515, 1996, 2665, 4744, 2031, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.32915854 0.08139846 0.41436577 ... 0.17102067 0.09254717 0.1166945 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2073, 2003, 1996, 4744, 2284, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.32915854 0.08139846 0.41436577 ... 0.17102067 0.09254717 0.1166945 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 4281, 1997, 2023, 6302, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.28426224 0.18149918 0.46163827 ... 0.13415167 0.2225182  0.10823505]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2097, 1996, 3902, 2022, 2437, 2151, 2062, 6762, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.28426224 0.18149918 0.46163827 ... 0.13415167 0.2225182  0.10823505]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 1037, 3902, 1999, 1996, 6302, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.28426224 0.18149918 0.46163827 ... 0.13415167 0.2225182  0.10823505]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2311, 2003, 2182, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.16677833 0.03810431 0.00069006 ... 0.09201534 0.05285855 0.63084483]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996, 27747,  4147,  6007,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.16677833 0.03810431 0.00069006 ... 0.09201534 0.05285855 0.63084483]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996,  4400, 12283,  2030,  7820,  2084,  1996,\n",
            "         2158,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.16677833 0.03810431 0.00069006 ... 0.09201534 0.05285855 0.63084483]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 2019, 2895, 2915, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.45650032 0.6020106  0.62203664 ... 1.3166382  0.72721255 1.0065091 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2024,  2122, 22054,  2005,  2336,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.45650032 0.6020106  0.62203664 ... 1.3166382  0.72721255 1.0065091 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 22054,  2024,  2182,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.45650032 0.6020106  0.62203664 ... 1.3166382  0.72721255 1.0065091 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2806, 2003, 2023, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.21211338 0.1857828  0.06780584 ... 0.09254242 0.3317977  1.3736703 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 2045, 2151, 2627, 5134, 2006, 1996, 2795, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.21211338 0.1857828  0.06780584 ... 0.09254242 0.3317977  1.3736703 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2097, 5572, 2022, 2366, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.21211338 0.1857828  0.06780584 ... 0.09254242 0.3317977  1.3736703 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996,  5572, 11008,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.21211338 0.1857828  0.06780584 ... 0.09254242 0.3317977  1.3736703 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2024,  1996,  2795, 23095,  2015,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06073769 0.09285713 0.46388632 ... 0.5001489  0.29111704 0.91734093]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 4916, 3902, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06073769 0.09285713 0.46388632 ... 0.5001489  0.29111704 0.91734093]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 4316, 2003, 2023, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06073769 0.09285713 0.46388632 ... 0.5001489  0.29111704 0.91734093]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 16), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2515, 1996, 2630, 2675, 6454, 2006, 1996, 2392, 1997,\n",
            "        1996, 3902, 2812, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 16), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.40691566 0.09357054 0.1509108  ... 0.07004023 0.68678    0.37377182]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 4111, 5777, 1999, 1037, 5430, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.40691566 0.09357054 0.1509108  ... 0.07004023 0.68678    0.37377182]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2828, 1997, 4111, 2003, 2023, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.40691566 0.09357054 0.1509108  ... 0.07004023 0.68678    0.37377182]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 9201, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.7299594  0.5489686  0.87457895 ... 0.19761373 1.3550119  1.1553792 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2515,  1996,  9850,  2298, 11937, 21756,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.7299594  0.5489686  0.87457895 ... 0.19761373 1.3550119  1.1553792 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2640, 2001, 1996, 9850, 2081, 2046, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.7299594  0.5489686  0.87457895 ... 0.19761373 1.3550119  1.1553792 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 9850, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.7299594  0.5489686  0.87457895 ... 0.19761373 1.3550119  1.1553792 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2038, 1996, 9850, 2042, 3013, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.7299594  0.5489686  0.87457895 ... 0.19761373 1.3550119  1.1553792 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  2016, 10261,  2005,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.7299594  0.5489686  0.87457895 ... 0.19761373 1.3550119  1.1553792 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2785,  1997, 16717,  2024,  6247,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.67628723 0.41662976 0.32072115 ... 0.5918906  0.2097159  0.3208638 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2064, 1996, 7997, 2175, 2006, 1996, 3902, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.67628723 0.41662976 0.32072115 ... 0.5918906  0.2097159  0.3208638 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 2845, 3345, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.67628723 0.41662976 0.32072115 ... 0.5918906  0.2097159  0.3208638 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2064, 2017, 3191, 2054, 1996, 3902, 3696, 2758, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.45794225 0.12824753 0.08733421 ... 0.3023339  0.10313624 0.09035148]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2024, 1996, 3210, 2006, 1996, 2346, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.45794225 0.12824753 0.08733421 ... 0.3023339  0.10313624 0.09035148]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 4874, 2003, 1999, 2392, 1997, 1996, 3696, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.45794225 0.12824753 0.08733421 ... 0.3023339  0.10313624 0.09035148]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3435, 2833, 4825, 2003, 1999, 1996, 3292, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.45794225 0.12824753 0.08733421 ... 0.3023339  0.10313624 0.09035148]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2024, 1996, 4534, 5697, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.45794225 0.12824753 0.08733421 ... 0.3023339  0.10313624 0.09035148]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2339,  2003,  4026,  2012,  1037,  4832, 28345,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.45794225 0.12824753 0.08733421 ... 0.3023339  0.10313624 0.09035148]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 4316, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.15119275 0.40133584 0.2522756  ... 0.38224903 0.14424723 0.43225113]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2172, 3384, 2003, 1999, 1996, 3861, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.15119275 0.40133584 0.2522756  ... 0.38224903 0.14424723 0.43225113]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 4742, 8131, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.15119275 0.40133584 0.2522756  ... 0.38224903 0.14424723 0.43225113]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 7444, 3609, 1999, 1996, 3746, 4281, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.02614614 0.02999703 0.38283646 ... 0.05849907 0.4077146  0.3539697 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 4825, 2024, 2027, 2012, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.02614614 0.02999703 0.38283646 ... 0.05849907 0.4077146  0.3539697 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2040, 2842, 2003, 1999, 1996, 9118, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.02614614 0.02999703 0.38283646 ... 0.05849907 0.4077146  0.3539697 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2079, 2122, 5194, 2031, 2006, 6397, 2545, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.02614614 0.02999703 0.38283646 ... 0.05849907 0.4077146  0.3539697 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 3586, 4815, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25853696 0.36376065 0.12852767 ... 0.40494972 0.311196   0.        ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 2158, 4755, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25853696 0.36376065 0.12852767 ... 0.40494972 0.311196   0.        ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2079,  1996, 12977,  2015,  2031,  1999,  2691,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25853696 0.36376065 0.12852767 ... 0.40494972 0.311196   0.        ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[  101,  2003,  2009, 24057,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25853696 0.36376065 0.12852767 ... 0.40494972 0.311196   0.        ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 4633, 2066, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25853696 0.36376065 0.12852767 ... 0.40494972 0.311196   0.        ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996, 12977,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25853696 0.36376065 0.12852767 ... 0.40494972 0.311196   0.        ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 3541, 4044, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25853696 0.36376065 0.12852767 ... 0.40494972 0.311196   0.        ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 4520, 1997, 3162, 2024, 1999, 1996, 2346, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.33140463 0.27043504 0.25006884 ... 0.26469946 0.22840956 0.1542241 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2008, 7997, 2835, 3730, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.33140463 0.27043504 0.25006884 ... 0.26469946 0.22840956 0.1542241 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996,  7997,  2108, 15230,  2006,  6900,  2030,\n",
            "        16295,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.33140463 0.27043504 0.25006884 ... 0.26469946 0.22840956 0.1542241 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 7700, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.33140463 0.27043504 0.25006884 ... 0.26469946 0.22840956 0.1542241 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2024, 1996, 2111, 1999, 4367, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.33140463 0.27043504 0.25006884 ... 0.26469946 0.22840956 0.1542241 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 22674,  2024, 15885,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.33140463 0.27043504 0.25006884 ... 0.26469946 0.22840956 0.1542241 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 2346, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.33140463 0.27043504 0.25006884 ... 0.26469946 0.22840956 0.1542241 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2097, 1996, 2111, 2991, 2091, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.33140463 0.27043504 0.25006884 ... 0.26469946 0.22840956 0.1542241 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 2111, 2024, 2006, 1996, 7997, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.33140463 0.27043504 0.25006884 ... 0.26469946 0.22840956 0.1542241 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609, 22674,  2024,  1996,  8195,  4147,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.33140463 0.27043504 0.25006884 ... 0.26469946 0.22840956 0.1542241 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 18105,  2024,  2045,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.40864637 0.2777846  0.84510946 ... 0.00407097 0.50521797 0.10941232]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2515,  1996,  4633,  3711, 16373,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.40864637 0.2777846  0.84510946 ... 0.00407097 0.50521797 0.10941232]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2024, 2027, 3788, 2058, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.40864637 0.2777846  0.84510946 ... 0.00407097 0.50521797 0.10941232]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996,  2158,  2383,  4390,  2007,  2010, 12977,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.38950986 0.02529645 0.33887583 ... 0.05743257 0.0111255  0.14746554]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996,  2158,  3061,  2030,  3564,  2006,  1996,\n",
            "        14175,  6277,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.38950986 0.02529645 0.33887583 ... 0.05743257 0.0111255  0.14746554]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 3712, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.38950986 0.02529645 0.33887583 ... 0.05743257 0.0111255  0.14746554]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 3609, 1997, 1996, 2300, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.46587145 0.11106516 0.80666304 ... 0.7202032  1.0516059  1.8007822 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2073, 2003, 1996, 3332, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.46587145 0.11106516 0.80666304 ... 0.7202032  1.0516059  1.8007822 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 6544, 4180, 1997, 2023, 4511, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.46587145 0.11106516 0.80666304 ... 0.7202032  1.0516059  1.8007822 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 4435, 2003, 4511, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.46587145 0.11106516 0.80666304 ... 0.7202032  1.0516059  1.8007822 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 2417, 4392, 2366, 1999, 2019, 6413, 3221, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.46587145 0.11106516 0.80666304 ... 0.7202032  1.0516059  1.8007822 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2106, 2017, 3046, 2023, 4511, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.43391377 0.27919084 0.4973259  ... 0.4175979  0.49135497 0.30852976]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2339, 2003, 2023, 2158, 3061, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.43391377 0.27919084 0.4973259  ... 0.4175979  0.49135497 0.30852976]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2071,  2045,  2022,  1037,  2304, 13383,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.43391377 0.27919084 0.4973259  ... 0.4175979  0.49135497 0.30852976]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 1016, 5167, 2008, 1996, 2158, 2003, 4147, 2674, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.43391377 0.27919084 0.4973259  ... 0.4175979  0.49135497 0.30852976]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3655, 2079, 2017, 2228, 2023, 3124, 7777, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.24824807 0.08464609 0.40953213 ... 0.09252398 0.07488388 0.00355188]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 2464, 2006, 1996, 9019, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.24824807 0.08464609 0.40953213 ... 0.09252398 0.07488388 0.00355188]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 5080, 2109, 2000, 2491, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.24824807 0.08464609 0.40953213 ... 0.09252398 0.07488388 0.00355188]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996,  8000, 11687,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.24824807 0.08464609 0.40953213 ... 0.09252398 0.07488388 0.00355188]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2064, 2017, 2113, 2073, 2017, 2024, 2478, 2023, 3042,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.24824807 0.08464609 0.40953213 ... 0.09252398 0.07488388 0.00355188]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2515,  2023,  9019,  2342,  2000,  2022,  2624, 25090,\n",
            "         5422,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.24824807 0.08464609 0.40953213 ... 0.09252398 0.07488388 0.00355188]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2023, 9019, 3214, 2005, 6001, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.24824807 0.08464609 0.40953213 ... 0.09252398 0.07488388 0.00355188]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 1037, 8000, 1999, 1996, 3861, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.4250217  0.2879606  0.31563354 ... 0.67523164 0.16412818 0.9410734 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 2023, 2019, 7344, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.4250217  0.2879606  0.31563354 ... 0.67523164 0.16412818 0.9410734 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2006, 1996, 2158, 1005, 1055, 2227, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.4250217  0.2879606  0.31563354 ... 0.67523164 0.16412818 0.9410734 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 2158, 2025, 4147, 6471, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.4250217  0.2879606  0.31563354 ... 0.67523164 0.16412818 0.9410734 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 6454, 2003, 2006, 1996, 2392, 1997, 1996, 6045, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.34058648 0.46099734 0.3997519  ... 0.59981084 1.528112   0.12395892]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2052, 2017, 4521, 2023, 2005, 6265, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.34058648 0.46099734 0.3997519  ... 0.59981084 1.528112   0.12395892]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 7965, 7954, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.34058648 0.46099734 0.3997519  ... 0.59981084 1.528112   0.12395892]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 5127, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.34058648 0.46099734 0.3997519  ... 0.59981084 1.528112   0.12395892]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2038,  2023, 11642,  2042,  8828,  2664,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.32804418 0.42691976 0.3161321  ... 0.6774022  0.45839214 0.3442544 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 2158, 4147, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.32804418 0.42691976 0.3161321  ... 0.6774022  0.45839214 0.3442544 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2073, 2003, 1996, 2158, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.32804418 0.42691976 0.3161321  ... 0.6774022  0.45839214 0.3442544 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2054,  4144,  2024,  2006,  1996, 14175,  6277,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.32804418 0.42691976 0.3161321  ... 0.6774022  0.45839214 0.3442544 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996, 14175,  6277,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6527935  0.9906112  0.7402878  ... 0.1892283  0.19246353 0.13458414]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 2450, 3173, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6527935  0.9906112  0.7402878  ... 0.1892283  0.19246353 0.13458414]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2016, 2559, 1999, 1037, 5259, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6527935  0.9906112  0.7402878  ... 0.1892283  0.19246353 0.13458414]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2014,  2132,  5058,  1999,  1037, 10257,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6527935  0.9906112  0.7402878  ... 0.1892283  0.19246353 0.13458414]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996,  2611,  4147, 11912,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6527935  0.9906112  0.7402878  ... 0.1892283  0.19246353 0.13458414]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996, 11868, 18623,  1999,  2014,  2677,  1996,\n",
            "         2157,  2126,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6527935  0.9906112  0.7402878  ... 0.1892283  0.19246353 0.13458414]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2106,  1996,  2450,  2079,  2077, 12766,  2014,\n",
            "         4091,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6527935  0.9906112  0.7402878  ... 0.1892283  0.19246353 0.13458414]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 2711, 3173, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.4167447  0.2835102  0.37019834 ... 0.81050706 0.2955023  0.59312344]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 5814, 6194, 2064, 2023, 4111, 2022, 2109, 2005, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.4167447  0.2835102  0.37019834 ... 0.81050706 0.2955023  0.59312344]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 8875, 2003, 2081, 2013, 2023, 4111, 1005, 1055, 6519,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.4167447  0.2835102  0.37019834 ... 0.81050706 0.2955023  0.59312344]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 2711, 7244, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.9180718  0.24831638 0.04590295 ... 0.12221237 0.23334473 0.31537825]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2024, 2122, 4111, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.9180718  0.24831638 0.04590295 ... 0.12221237 0.23334473 0.31537825]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 2035, 2122, 4176, 5307, 1996, 2168, 3257, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.9180718  0.24831638 0.04590295 ... 0.12221237 0.23334473 0.31537825]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 21025, 27528,  7959,  2015,  2024,  2045,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.9180718  0.24831638 0.04590295 ... 0.12221237 0.23334473 0.31537825]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2073,  2024,  1996, 21025, 27528,  7959,  2015,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.9180718  0.24831638 0.04590295 ... 0.12221237 0.23334473 0.31537825]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 2915, 1999, 3609, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.9180718  0.24831638 0.04590295 ... 0.12221237 0.23334473 0.31537825]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2024, 1996, 4176, 2770, 2015, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.9180718  0.24831638 0.04590295 ... 0.12221237 0.23334473 0.31537825]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2024, 2045, 2151, 3628, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.9180718  0.24831638 0.04590295 ... 0.12221237 0.23334473 0.31537825]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 4111, 2003, 1999, 1996, 3861, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.9180718  0.24831638 0.04590295 ... 0.12221237 0.23334473 0.31537825]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 16), dtype=int32, numpy=\n",
            "array([[ 101, 2612, 1997, 3173, 2398, 2054, 2064, 2122, 4176, 2907, 2043,\n",
            "        2027, 2693, 2362, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 16), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.9180718  0.24831638 0.04590295 ... 0.12221237 0.23334473 0.31537825]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2079, 2122, 4176, 4521, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.9180718  0.24831638 0.04590295 ... 0.12221237 0.23334473 0.31537825]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2073, 2003, 1996, 5568, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.9180718  0.24831638 0.04590295 ... 0.12221237 0.23334473 0.31537825]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 4176, 2024, 3491, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04206308 0.05598668 0.13848552 ... 0.25683737 0.38574773 0.6691063 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  1996, 10777,  3173,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04206308 0.05598668 0.13848552 ... 0.25683737 0.38574773 0.6691063 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2029,  4874,  2003,  4487, 13102, 18981, 11589, 19798,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04206308 0.05598668 0.13848552 ... 0.25683737 0.38574773 0.6691063 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2097,  1996, 10777,  2022, 13487,  2000,  2079,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04206308 0.05598668 0.13848552 ... 0.25683737 0.38574773 0.6691063 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 3522, 6302, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04206308 0.05598668 0.13848552 ... 0.25683737 0.38574773 0.6691063 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  1996, 10777,  6183,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04206308 0.05598668 0.13848552 ... 0.25683737 0.38574773 0.6691063 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 4111, 2003, 3491, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04206308 0.05598668 0.13848552 ... 0.25683737 0.38574773 0.6691063 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 2111, 2024, 3564, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04206308 0.05598668 0.13848552 ... 0.25683737 0.38574773 0.6691063 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 4111, 2003, 2023, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2039671  0.1341866  0.28086767 ... 0.72158813 0.13077486 0.08743805]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2024, 1996, 4597, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2039671  0.1341866  0.28086767 ... 0.72158813 0.13077486 0.08743805]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 4597, 2024, 2045, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2039671  0.1341866  0.28086767 ... 0.72158813 0.13077486 0.08743805]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 2045, 2151, 8044, 1999, 1996, 3712, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1194209  0.2364504  0.18470265 ... 0.48306537 0.5370682  0.61365646]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2009, 1037, 4010, 2154, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1194209  0.2364504  0.18470265 ... 0.48306537 0.5370682  0.61365646]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 2158, 1999, 1996, 6302, 3201, 2000, 8301, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1194209  0.2364504  0.18470265 ... 0.48306537 0.5370682  0.61365646]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2038,  1996,  4586,  2042, 18087,  2098,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1194209  0.2364504  0.18470265 ... 0.48306537 0.5370682  0.61365646]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2009, 4586, 2075, 3805, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1194209  0.2364504  0.18470265 ... 0.48306537 0.5370682  0.61365646]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 4987, 2000, 2010, 2519, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1194209  0.2364504  0.18470265 ... 0.48306537 0.5370682  0.61365646]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 2009, 4586, 2075, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1194209  0.2364504  0.18470265 ... 0.48306537 0.5370682  0.61365646]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2001,  2023,  6302,  2579,  1999,  1037,  4586, 19718,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1194209  0.2364504  0.18470265 ... 0.48306537 0.5370682  0.61365646]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2024, 1996, 2111, 2725, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3941628  0.18413891 0.24815638 ... 0.44244164 0.17650999 0.00966548]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 4562, 2003, 2023, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3941628  0.18413891 0.24815638 ... 0.44244164 0.17650999 0.00966548]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 3748, 4562, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3941628  0.18413891 0.24815638 ... 0.44244164 0.17650999 0.00966548]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 4605, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.19798309 0.28475803 0.6823077  ... 0.19415131 0.0724102  0.2409994 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 2158, 3173, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.19798309 0.28475803 0.6823077  ... 0.19415131 0.0724102  0.2409994 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2040, 2003, 2008, 2158, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.19798309 0.28475803 0.6823077  ... 0.19415131 0.0724102  0.2409994 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2073, 2003, 2023, 2173, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.19798309 0.28475803 0.6823077  ... 0.19415131 0.0724102  0.2409994 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2040,  2038,  1037, 17983,  3797,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.19798309 0.28475803 0.6823077  ... 0.19415131 0.0724102  0.2409994 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[  101,  2024,  2122, 10439, 20624,  6290,  1005,  1055,  1999,\n",
            "         1037,  4825,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.19798309 0.28475803 0.6823077  ... 0.19415131 0.0724102  0.2409994 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2073,  2003,  1996, 15243,  4094,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.845422   0.11158554 0.3773954  ... 0.01719695 0.95589745 0.95792425]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1999, 2030, 2648, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.845422   0.11158554 0.3773954  ... 0.01719695 0.95589745 0.95792425]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2515,  2003,  2360,  2006,  2216, 16143,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.845422   0.11158554 0.3773954  ... 0.01719695 0.95589745 0.95792425]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2003, 2108, 6913, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06333753 0.18094373 0.05234423 ... 0.27680948 0.21832433 0.30856392]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2003, 2002, 4147, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06333753 0.18094373 0.05234423 ... 0.27680948 0.21832433 0.30856392]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2008,  1037, 13137,  6045,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06333753 0.18094373 0.05234423 ... 0.27680948 0.21832433 0.30856392]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2001,  2023,  3861,  2081,  2006,  1037, 11559,  2154,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06333753 0.18094373 0.05234423 ... 0.27680948 0.21832433 0.30856392]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2001, 2023, 2579, 1999, 1996, 3438, 1005, 1055, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06333753 0.18094373 0.05234423 ... 0.27680948 0.21832433 0.30856392]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 5495, 2515, 1996, 2158, 2031, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06333753 0.18094373 0.05234423 ... 0.27680948 0.21832433 0.30856392]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 2158, 1005, 1055, 6598, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06333753 0.18094373 0.05234423 ... 0.27680948 0.21832433 0.30856392]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 2006, 2010, 2132, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06333753 0.18094373 0.05234423 ... 0.27680948 0.21832433 0.30856392]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 4848, 2003, 2008, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.168326   0.2918691  0.2679854  ... 0.06515481 0.2096531  0.4879877 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2003,  1996, 10171,  2613,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.168326   0.2918691  0.2679854  ... 0.06515481 0.2096531  0.4879877 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2073, 2024, 2027, 5825, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.168326   0.2918691  0.2679854  ... 0.06515481 0.2096531  0.4879877 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 4049, 2003, 2008, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08975329 0.5140819  0.41056016 ... 1.0240209  0.8576833  0.0553703 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 2711, 3173, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08975329 0.5140819  0.41056016 ... 1.0240209  0.8576833  0.0553703 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2515,  1996, 20497,  3711,  2000,  2022,  2081,\n",
            "         2041,  1997,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08975329 0.5140819  0.41056016 ... 1.0240209  0.8576833  0.0553703 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 6302, 9543, 3436, 6028, 2001, 4162, 2000, 2023, 3861,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08975329 0.5140819  0.41056016 ... 1.0240209  0.8576833  0.0553703 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 3124, 3173, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.51873755 0.8415673  0.48510462 ... 0.38242793 0.9074248  0.02587281]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2097, 2002, 3613, 2000, 2991, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.51873755 0.8415673  0.48510462 ... 0.38242793 0.9074248  0.02587281]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2024,  2045,  2151, 13137, 20968,  2006,  1996,  9850,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.51873755 0.8415673  0.48510462 ... 0.38242793 0.9074248  0.02587281]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2214, 2003, 2023, 2158, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.59019965 0.16072685 0.7743982  ... 0.10375541 0.6018143  0.5980631 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 2009, 4586, 2075, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.59019965 0.16072685 0.7743982  ... 0.10375541 0.6018143  0.5980631 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2024, 2122, 3748, 4176, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.59019965 0.16072685 0.7743982  ... 0.10375541 0.6018143  0.5980631 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2003, 2023, 4111, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.59019965 0.16072685 0.7743982  ... 0.10375541 0.6018143  0.5980631 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023, 24274,  2030, 19350,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06654052 1.1813087  0.00304021 ... 0.82357526 0.4927771  0.45697135]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116,  4109,  1997, 10733,  2024,  2006,  1996,\n",
            "         5127,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06654052 1.1813087  0.00304021 ... 0.82357526 0.4927771  0.45697135]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2064, 2017, 4521, 2023, 2007, 2115, 2398, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06654052 1.1813087  0.00304021 ... 0.82357526 0.4927771  0.45697135]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[  101,  2001,  2023, 10733, 12984,  1999,  1037,  3536,  1011,\n",
            "         5255, 17428,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06302395 0.22198854 0.07237547 ... 0.3939643  0.3567473  0.6225682 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 2111, 2024, 1999, 1996, 3861, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06302395 0.22198854 0.07237547 ... 0.3939643  0.3567473  0.6225682 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2024, 1996, 2048, 2273, 2725, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06302395 0.22198854 0.07237547 ... 0.3939643  0.3567473  0.6225682 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023,  1037, 24067,  2100,  2828,  2154,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06302395 0.22198854 0.07237547 ... 0.3939643  0.3567473  0.6225682 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023,  8301,  5376,  2158, 21565,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06302395 0.22198854 0.07237547 ... 0.3939643  0.3567473  0.6225682 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 2111, 2024, 1999, 1996, 2250, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06302395 0.22198854 0.07237547 ... 0.3939643  0.3567473  0.6225682 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2040, 2003, 1999, 1996, 2250, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06302395 0.22198854 0.07237547 ... 0.3939643  0.3567473  0.6225682 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996, 21294,  2183,  2000,  2991,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06302395 0.22198854 0.07237547 ... 0.3939643  0.3567473  0.6225682 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2024,  1996,  2111,  2731,  2005,  1037, 12701,  2977,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.45874494 0.145805   0.10634892 ... 0.02658098 0.26873884 0.26130885]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2214, 2003, 1996, 3899, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.45874494 0.145805   0.10634892 ... 0.02658098 0.26873884 0.26130885]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 3899, 2006, 1037, 2346, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.45874494 0.145805   0.10634892 ... 0.02658098 0.26873884 0.26130885]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 9785, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.45874494 0.145805   0.10634892 ... 0.02658098 0.26873884 0.26130885]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2106,  1996,  3899,  2074,  4608,  1037, 10424,  2483,\n",
            "        11306,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.45874494 0.145805   0.10634892 ... 0.02658098 0.26873884 0.26130885]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996,  3899,  2183,  2000,  2507,  1996, 10424,\n",
            "         2483, 11306,  2067,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.45874494 0.145805   0.10634892 ... 0.02658098 0.26873884 0.26130885]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 2158, 1999, 1037, 2380, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5650666  0.21369348 0.34313715 ... 1.2262485  0.5712869  0.097304  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2024, 2035, 1996, 3121, 2317, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5650666  0.21369348 0.34313715 ... 1.2262485  0.5712869  0.097304  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 1005, 3788, 1005, 2422, 1999, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5650666  0.21369348 0.34313715 ... 1.2262485  0.5712869  0.097304  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 5751, 2024, 1999, 2023, 3496, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5650666  0.21369348 0.34313715 ... 1.2262485  0.5712869  0.097304  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 3609, 1997, 1996, 2644, 7138, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5650666  0.21369348 0.34313715 ... 1.2262485  0.5712869  0.097304  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 15), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2009,  3647,  2005,  2111,  2000,  3328,  2408,\n",
            "         1996,  2892, 17122,  2085,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 15), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5650666  0.21369348 0.34313715 ... 1.2262485  0.5712869  0.097304  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2064, 24946,  3328,  2408,  7063,  2358,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5650666  0.21369348 0.34313715 ... 1.2262485  0.5712869  0.097304  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2126, 2003, 1996, 2028, 2126, 3696, 7302, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5650666  0.21369348 0.34313715 ... 1.2262485  0.5712869  0.097304  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2029, 2126, 2515, 1996, 8612, 2391, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11495055 0.14115652 0.36469096 ... 0.00557434 0.7986637  0.02407189]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023,  2019,  6034, 17260,  6277,  2121,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11495055 0.14115652 0.36469096 ... 0.00557434 0.7986637  0.02407189]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3019, 5783, 2003, 1999, 1996, 3292, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11495055 0.14115652 0.36469096 ... 0.00557434 0.7986637  0.02407189]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2051, 1997, 2154, 2003, 2009, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11495055 0.14115652 0.36469096 ... 0.00557434 0.7986637  0.02407189]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2073, 2003, 2023, 2158, 3564, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11495055 0.14115652 0.36469096 ... 0.00557434 0.7986637  0.02407189]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023,  2711,  5102, 23263,  2005,  1996,  2154,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11495055 0.14115652 0.36469096 ... 0.00557434 0.7986637  0.02407189]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2515,  2023,  2711,  5959,  4654, 26415,  8156,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11495055 0.14115652 0.36469096 ... 0.00557434 0.7986637  0.02407189]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 5749, 2024, 2045, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11495055 0.14115652 0.36469096 ... 0.00557434 0.7986637  0.02407189]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 3124, 3564, 2006, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11495055 0.14115652 0.36469096 ... 0.00557434 0.7986637  0.02407189]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[  101,  2515,  2023,  2711,  3711,  2000,  2022, 29156,  1999,\n",
            "         1996,  9325,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11495055 0.14115652 0.36469096 ... 0.00557434 0.7986637  0.02407189]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2158, 3173, 2006, 2010, 5001, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11495055 0.14115652 0.36469096 ... 0.00557434 0.7986637  0.02407189]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 1037, 2502, 2103, 2369, 1996, 2158, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.05557206 0.21222909 0.28083843 ... 0.17540233 0.60325974 0.19136384]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 7997, 8195, 2064, 2022, 2464, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.05557206 0.21222909 0.28083843 ... 0.17540233 0.60325974 0.19136384]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 2023, 2711, 3147, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.05557206 0.21222909 0.28083843 ... 0.17540233 0.60325974 0.19136384]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2073, 2003, 1037, 2540, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.05557206 0.21222909 0.28083843 ... 0.17540233 0.60325974 0.19136384]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 6077, 2024, 1999, 2023, 3746, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.05557206 0.21222909 0.28083843 ... 0.17540233 0.60325974 0.19136384]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2003,  2008,  1037, 17763,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.05557206 0.21222909 0.28083843 ... 0.17540233 0.60325974 0.19136384]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2515,  2023,  2711,  2031, 13967,  2006,  2014,  2303,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.17969197 0.5076378  0.02211301 ... 1.2356365  0.43496126 0.07589899]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2828, 1997, 8808, 2003, 2008, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.17969197 0.5076378  0.02211301 ... 1.2356365  0.43496126 0.07589899]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2038,  2023,  2833,  2042, 12984,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.17969197 0.5076378  0.02211301 ... 1.2356365  0.43496126 0.07589899]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2052,  1037, 15942,  2078,  4521,  2023,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.17969197 0.5076378  0.02211301 ... 1.2356365  0.43496126 0.07589899]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 2833, 2003, 1999, 2023, 6090, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.50176823 0.15281151 0.23076427 ... 0.47536665 0.3101738  0.2977041 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2158, 1999, 2023, 6302, 2725, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.50176823 0.15281151 0.23076427 ... 0.47536665 0.3101738  0.2977041 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2158, 2725, 1999, 1996, 4153, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.50176823 0.15281151 0.23076427 ... 0.47536665 0.3101738  0.2977041 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 2158, 2725, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.50176823 0.15281151 0.23076427 ... 0.47536665 0.3101738  0.2977041 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2023,  4368,  2003,  4050,  2641, 12320,  2030, 14818,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.50176823 0.15281151 0.23076427 ... 0.47536665 0.3101738  0.2977041 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 4368, 2003, 1996, 8258, 8019, 1999, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.50176823 0.15281151 0.23076427 ... 0.47536665 0.3101738  0.2977041 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996, 11498, 15816,  2140,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.50176823 0.15281151 0.23076427 ... 0.47536665 0.3101738  0.2977041 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 2111, 2024, 1999, 1996, 2300, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5831636  0.37141407 0.5593772  ... 0.19026819 0.39019883 0.66610944]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2001,  2023,  9850,  2081,  2012,  1037, 18112,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5831636  0.37141407 0.5593772  ... 0.19026819 0.39019883 0.66610944]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054, 11446,  2003,  2006,  2327,  1997,  1996,  9850,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5831636  0.37141407 0.5593772  ... 0.19026819 0.39019883 0.66610944]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 2795, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5831636  0.37141407 0.5593772  ... 0.19026819 0.39019883 0.66610944]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2003,  2023,  2005, 18064,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.45946592 0.36093315 0.49618608 ... 0.06878255 0.6824672  0.3657537 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 9639, 1997, 5572, 2024, 3491, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.45946592 0.36093315 0.49618608 ... 0.06878255 0.6824672  0.3657537 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996, 10810,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.45946592 0.36093315 0.49618608 ... 0.06878255 0.6824672  0.3657537 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 16), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2828,  1997,  6381,  2024,  1999,  1996, 25234,\n",
            "        26536,  2015,  2006,  1996,  2795,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 16), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.45946592 0.36093315 0.49618608 ... 0.06878255 0.6824672  0.3657537 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 4653, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.10959148 0.08680935 0.31820804 ... 0.24102266 0.18487775 0.49767512]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[  101,  2003,  2023, 24274,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.10959148 0.08680935 0.31820804 ... 0.24102266 0.18487775 0.49767512]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2024,  2045, 24213,  2006,  1996,  2723,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.10959148 0.08680935 0.31820804 ... 0.24102266 0.18487775 0.49767512]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 2542, 2282, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.10959148 0.08680935 0.31820804 ... 0.24102266 0.18487775 0.49767512]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2024,  2216,  6116, 20053,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.10959148 0.08680935 0.31820804 ... 0.24102266 0.18487775 0.49767512]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996, 10257,  2006,  1996, 14366,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.10959148 0.08680935 0.31820804 ... 0.24102266 0.18487775 0.49767512]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2023, 2282, 2109, 2005, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.10959148 0.08680935 0.31820804 ... 0.24102266 0.18487775 0.49767512]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2828,  1997,  7497,  2003, 27816,  2023,  7169,\n",
            "         2686,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.10959148 0.08680935 0.31820804 ... 0.24102266 0.18487775 0.49767512]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 7497, 2001, 5361, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.10959148 0.08680935 0.31820804 ... 0.24102266 0.18487775 0.49767512]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 2023, 5723, 2736, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.10959148 0.08680935 0.31820804 ... 0.24102266 0.18487775 0.49767512]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2029, 2282, 2003, 2023, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.10959148 0.08680935 0.31820804 ... 0.24102266 0.18487775 0.49767512]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116,  4597, 17407,  2024,  2045,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.29778552 0.6385563  0.3732179  ... 0.891637   0.81310296 0.49860552]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[ 101, 2003, 2009, 3147, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.29778552 0.6385563  0.3732179  ... 0.891637   0.81310296 0.49860552]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[  101,  2024,  2027, 17662,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.29778552 0.6385563  0.3732179  ... 0.891637   0.81310296 0.49860552]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 1037, 4946, 1999, 1996, 2250, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.29778552 0.6385563  0.3732179  ... 0.891637   0.81310296 0.49860552]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 20), dtype=int32, numpy=\n",
            "array([[  101,  2515,  2023,  2711,  2031,  1037,  2488,  2084,  2779,\n",
            "         3382,  1997,  6114,  1037, 24665,  2666,  6767,  2271,  4544,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 20), dtype=int32, numpy=\n",
            "array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]],\n",
            "      dtype=int32)>}\n",
            "[0.29778552 0.6385563  0.3732179  ... 0.891637   0.81310296 0.49860552]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 2158, 2725, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.29778552 0.6385563  0.3732179  ... 0.891637   0.81310296 0.49860552]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1999, 1996, 3712, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.29778552 0.6385563  0.3732179  ... 0.891637   0.81310296 0.49860552]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2024,  2045,  5055, 17335,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.29778552 0.6385563  0.3732179  ... 0.891637   0.81310296 0.49860552]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[ 101, 2024, 2027, 3141, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.29778552 0.6385563  0.3732179  ... 0.891637   0.81310296 0.49860552]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1999, 1996, 2250, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.29778552 0.6385563  0.3732179  ... 0.891637   0.81310296 0.49860552]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 2111, 2024, 1999, 1996, 3861, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.29778552 0.6385563  0.3732179  ... 0.891637   0.81310296 0.49860552]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 2450, 2006, 1996, 2187, 4147, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5017769  0.07019702 0.19626379 ... 0.         0.30028802 0.60583276]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  2006,  1996, 11142,  2503,  1996,  3573,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5017769  0.07019702 0.19626379 ... 0.         0.30028802 0.60583276]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2773, 2006, 1996, 3221, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5017769  0.07019702 0.19626379 ... 0.         0.30028802 0.60583276]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 2045, 1037, 9185, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01823848 0.13261752 0.43419695 ... 0.41290817 0.01506272 0.04131402]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2828, 1997, 4633, 2003, 2023, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01823848 0.13261752 0.43419695 ... 0.41290817 0.01506272 0.04131402]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 2023, 2711, 2725, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01823848 0.13261752 0.43419695 ... 0.41290817 0.01506272 0.04131402]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023,  2711,  1037,  2658, 21294,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01823848 0.13261752 0.43419695 ... 0.41290817 0.01506272 0.04131402]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996,  2879,  4147, 11875,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5398766  0.4203117  1.6005291  ... 0.28160346 0.96347463 0.77451146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2024, 1996, 4870, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5398766  0.4203117  1.6005291  ... 0.28160346 0.96347463 0.77451146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 1996, 2111, 3564, 2030, 3061, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5398766  0.4203117  1.6005291  ... 0.28160346 0.96347463 0.77451146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 6686, 2515, 2023, 3711, 2000, 2022, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5398766  0.4203117  1.6005291  ... 0.28160346 0.96347463 0.77451146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 2111, 2024, 2045, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.48089993 0.63773537 0.26688898 ... 0.03035006 0.96132034 0.32076597]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 2813, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.48089993 0.63773537 0.26688898 ... 0.03035006 0.96132034 0.32076597]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2054,  2833,  2003, 15885,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.48089993 0.63773537 0.26688898 ... 0.03035006 0.96132034 0.32076597]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116,  7760,  2024,  1999,  2023,  8902, 20679,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11379439 0.377284   0.38704863 ... 0.12350109 0.97272825 0.00782061]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 2775, 3173, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11379439 0.377284   0.38704863 ... 0.12350109 0.97272825 0.00782061]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 2367, 4620, 2024, 2182, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11379439 0.377284   0.38704863 ... 0.12350109 0.97272825 0.00782061]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2024, 1996, 2048, 4871, 7235, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11379439 0.377284   0.38704863 ... 0.12350109 0.97272825 0.00782061]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  3564,  2006,  2327,  1997,  1996, 21290,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.03779417 0.10473365 0.0235131  ... 0.21660233 0.61105907 0.04493076]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2038,  2619,  2012,  2431,  1997,  1996, 11642,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.03779417 0.10473365 0.0235131  ... 0.21660233 0.61105907 0.04493076]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2003, 2023, 2833, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.03779417 0.10473365 0.0235131  ... 0.21660233 0.61105907 0.04493076]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2023, 2828, 1997, 7852, 2170, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.03779417 0.10473365 0.0235131  ... 0.21660233 0.61105907 0.04493076]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023,  1037,  2878, 11642,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.03779417 0.10473365 0.0235131  ... 0.21660233 0.61105907 0.04493076]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023,  1037,  4524,  2884, 11642,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.34162137 0.6666837  0.5850619  ... 0.89596486 0.19397733 0.38588047]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 20), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2744,  5577,  1037,  2022, 16416,  7178,  2564,\n",
            "         1998,  1996,  3328,  2012,  1996,  2327,  1997,  2023,  2311,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 20), dtype=int32, numpy=\n",
            "array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]],\n",
            "      dtype=int32)>}\n",
            "[0.34162137 0.6666837  0.5850619  ... 0.89596486 0.19397733 0.38588047]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 5119, 3418, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.34162137 0.6666837  0.5850619  ... 0.89596486 0.19397733 0.38588047]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=\n",
            "array([[ 101, 2171, 1996, 6549, 2806, 1997, 1996, 2311, 4637, 1996, 5119,\n",
            "        3578, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.34162137 0.6666837  0.5850619  ... 0.89596486 0.19397733 0.38588047]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 5119, 2551, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.34162137 0.6666837  0.5850619  ... 0.89596486 0.19397733 0.38588047]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2006, 2327, 1997, 1996, 5119, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.34162137 0.6666837  0.5850619  ... 0.89596486 0.19397733 0.38588047]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2051, 2003, 2009, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.22949435 0.39825055 0.30514526 ... 0.19905968 0.6684946  0.36806494]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  2028,  2518,  2006,  1996, 19475,  2795,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.22949435 0.39825055 0.30514526 ... 0.19905968 0.6684946  0.36806494]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2785, 1997, 3899, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.22949435 0.39825055 0.30514526 ... 0.19905968 0.6684946  0.36806494]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 8768, 1996, 2158, 2003, 2104, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18133815 0.18350539 0.35011882 ... 0.21460238 0.51189435 0.22883049]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 2158, 3407, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18133815 0.18350539 0.35011882 ... 0.21460238 0.51189435 0.22883049]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2006, 1996, 2158, 1005, 1055, 3797, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18133815 0.18350539 0.35011882 ... 0.21460238 0.51189435 0.22883049]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 2158, 2725, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18133815 0.18350539 0.35011882 ... 0.21460238 0.51189435 0.22883049]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 2158, 3173, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18133815 0.18350539 0.35011882 ... 0.21460238 0.51189435 0.22883049]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 2658, 8258, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18133815 0.18350539 0.35011882 ... 0.21460238 0.51189435 0.22883049]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2001, 2023, 6302, 2579, 2503, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18133815 0.18350539 0.35011882 ... 0.21460238 0.51189435 0.22883049]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 4715, 3608, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[7.0355272e-01 4.6345285e-01 1.1855383e-01 ... 3.1891749e-03 5.0569761e-01\n",
            " 2.4875475e-04]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 4597, 2079, 2017, 2156, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[7.0355272e-01 4.6345285e-01 1.1855383e-01 ... 3.1891749e-03 5.0569761e-01\n",
            " 2.4875475e-04]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2009, 12217,  2030, 26031,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[7.0355272e-01 4.6345285e-01 1.1855383e-01 ... 3.1891749e-03 5.0569761e-01\n",
            " 2.4875475e-04]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 4026, 2422, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.43843764 0.08630891 0.6899328  ... 0.06273544 1.3169699  0.08939541]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 1996, 3057, 5102, 1996, 2168, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.43843764 0.08630891 0.6899328  ... 0.06273544 1.3169699  0.08939541]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 2014, 2606, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.43843764 0.08630891 0.6899328  ... 0.06273544 1.3169699  0.08939541]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2064, 2122, 2336, 3191, 2394, 2006, 2037, 7588, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.43843764 0.08630891 0.6899328  ... 0.06273544 1.3169699  0.08939541]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 8080, 2006, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.43843764 0.08630891 0.6899328  ... 0.06273544 1.3169699  0.08939541]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2785, 1997, 3274, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.43843764 0.08630891 0.6899328  ... 0.06273544 1.3169699  0.08939541]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2008,  1037,  2204,  2173,  2005,  1037, 12191,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.43843764 0.08630891 0.6899328  ... 0.06273544 1.3169699  0.08939541]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 14804,  4244,  2024,  5689,  2006,  1996,\n",
            "         2813,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.43843764 0.08630891 0.6899328  ... 0.06273544 1.3169699  0.08939541]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2450, 1999, 1996, 5061, 3797, 2725, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.43843764 0.08630891 0.6899328  ... 0.06273544 1.3169699  0.08939541]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2006, 1996, 2187, 2217, 1997, 1996, 2795, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3774631  0.5072036  1.6327444  ... 0.17191987 0.5037638  0.7035475 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2024,  1996, 14006,  5507,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3774631  0.5072036  1.6327444  ... 0.17191987 0.5037638  0.7035475 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2785, 1997, 9850, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3774631  0.5072036  1.6327444  ... 0.17191987 0.5037638  0.7035475 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054, 24582,  2075,  2003,  2109,  2006,  2023,  9850,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20582114 0.41478792 0.44927263 ... 0.2605542  0.21880256 1.0675793 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 2679, 3586, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20582114 0.41478792 0.44927263 ... 0.2605542  0.21880256 1.0675793 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[ 101, 2129, 2116, 5194, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20582114 0.41478792 0.44927263 ... 0.2605542  0.21880256 1.0675793 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 3586, 1005, 1055, 5725, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20582114 0.41478792 0.44927263 ... 0.2605542  0.21880256 1.0675793 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 3586, 1999, 1996, 3224, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20582114 0.41478792 0.44927263 ... 0.2605542  0.21880256 1.0675793 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 1996, 5194, 1996, 2168, 8843, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20582114 0.41478792 0.44927263 ... 0.2605542  0.21880256 1.0675793 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 3586, 2725, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1957918  0.45080024 0.03065896 ... 0.2658807  0.03456178 0.87966835]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2024, 1996, 4597, 2006, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1957918  0.45080024 0.03065896 ... 0.2658807  0.03456178 0.87966835]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023,  2282,  1999,  1037, 21538,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1957918  0.45080024 0.03065896 ... 0.2658807  0.03456178 0.87966835]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2282, 2003, 2023, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.03047981 0.31346098 0.21530746 ... 1.6566921  0.11166034 0.19831854]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 2715, 2694, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.03047981 0.31346098 0.21530746 ... 1.6566921  0.11166034 0.19831854]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 2813, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.03047981 0.31346098 0.21530746 ... 1.6566921  0.11166034 0.19831854]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 3898, 2006, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.14885756 0.27468255 0.41270757 ... 0.75930077 0.27954292 0.67695796]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 2023, 2711, 2725, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.14885756 0.27468255 0.41270757 ... 0.75930077 0.27954292 0.67695796]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 6087, 2024, 1996, 2813, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.14885756 0.27468255 0.41270757 ... 0.75930077 0.27954292 0.67695796]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 2006, 1996, 2598, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.35740036 0.36468276 0.93985045 ... 0.741212   0.558151   0.36512232]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 3628, 2024, 2216, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.35740036 0.36468276 0.93985045 ... 0.741212   0.558151   0.36512232]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116,  2111,  2031, 16717,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.35740036 0.36468276 0.93985045 ... 0.741212   0.558151   0.36512232]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2024,  2070,  1997,  1996,  2111,  4147, 15806,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.35740036 0.36468276 0.93985045 ... 0.741212   0.558151   0.36512232]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2040,  2024,  2006,  1996, 18105,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08003967 0.18015297 0.44660637 ... 0.6263783  0.4429442  0.00286419]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2024,  1996, 23999,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08003967 0.18015297 0.44660637 ... 0.6263783  0.4429442  0.00286419]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 10424,  2483, 11306,  2015,  2024,  2045,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08003967 0.18015297 0.44660637 ... 0.6263783  0.4429442  0.00286419]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2106, 1996, 2158, 2991, 2091, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.98817015 0.4684439  0.2463012  ... 0.7113975  0.51881766 0.09308478]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023, 24274,  2030, 19350,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.98817015 0.4684439  0.2463012  ... 0.7113975  0.51881766 0.09308478]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 2014, 2606, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.98817015 0.4684439  0.2463012  ... 0.7113975  0.51881766 0.09308478]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  1996,  2611, 17492,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.98817015 0.4684439  0.2463012  ... 0.7113975  0.51881766 0.09308478]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 2833, 8875, 2016, 1005, 1055, 5983,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18211988 0.08600248 0.6657685  ... 0.40880516 0.67863154 0.15079138]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 2044, 3861, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18211988 0.08600248 0.6657685  ... 0.40880516 0.67863154 0.15079138]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 4734, 2003, 2023, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18211988 0.08600248 0.6657685  ... 0.40880516 0.67863154 0.15079138]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 1037, 2694, 1999, 1996, 3861, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20767264 0.32717687 0.25830874 ... 0.00419063 0.3059097  1.8026353 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 1037, 2304, 3586, 4089, 2464, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20767264 0.32717687 0.25830874 ... 0.00419063 0.3059097  1.8026353 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2024, 2045, 2116, 3628, 2182, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20767264 0.32717687 0.25830874 ... 0.00419063 0.3059097  1.8026353 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2339, 2001, 2023, 3861, 2579, 2061, 2521, 2185, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.8917955  0.23776813 1.0468836  ... 0.3982705  0.11940883 0.3455044 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[  101,  2003,  2009, 12217,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.8917955  0.23776813 1.0468836  ... 0.3982705  0.11940883 0.3455044 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2339, 4995, 1005, 1056, 2111, 1999, 1996, 4534, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.8917955  0.23776813 1.0468836  ... 0.3982705  0.11940883 0.3455044 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2073,  2003,  1996, 26018,  3372,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.46515924 0.12033728 0.15913676 ... 0.39992428 0.20997548 0.9219311 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2339,  2003,  1996, 18302,  2061,  2485,  2000,  1996,\n",
            "         7752,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.46515924 0.12033728 0.15913676 ... 0.39992428 0.20997548 0.9219311 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2282, 2003, 2023, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.46515924 0.12033728 0.15913676 ... 0.39992428 0.20997548 0.9219311 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 2282, 4550, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04249235 0.9788997  0.45564094 ... 0.8249978  0.45139867 0.47433862]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2158, 4147, 2006, 2010, 2849, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04249235 0.9788997  0.45564094 ... 0.8249978  0.45139867 0.47433862]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  9530, 21341,  2003,  2006,  1996,  2980,  3899,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04249235 0.9788997  0.45564094 ... 0.8249978  0.45139867 0.47433862]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2001,  2023,  3861,  2579,  2012,  1037, 10356,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04249235 0.9788997  0.45564094 ... 0.8249978  0.45139867 0.47433862]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 2711, 3173, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.27523196 0.09488107 0.04290161 ... 0.0601258  0.17229547 0.13020012]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 2045, 2300, 2182, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.27523196 0.09488107 0.04290161 ... 0.0601258  0.17229547 0.13020012]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 4743, 2003, 2023, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.27523196 0.09488107 0.04290161 ... 0.0601258  0.17229547 0.13020012]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996,  4743,  1005,  1055, 23525,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.27523196 0.09488107 0.04290161 ... 0.0601258  0.17229547 0.13020012]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 5980, 2015, 2515, 1996, 4743, 2031, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.49519998 0.19032954 0.21565199 ... 0.14546299 0.38027892 0.11843546]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[  101,  2003,  2009, 24057,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.49519998 0.19032954 0.21565199 ... 0.14546299 0.38027892 0.11843546]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 2346, 5697, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.49519998 0.19032954 0.21565199 ... 0.14546299 0.38027892 0.11843546]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2339, 2024, 1996, 2111, 2770, 2408, 1996, 2395, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.341878   0.35559386 0.73415124 ... 0.14034525 0.05180654 0.757296  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 5581, 8316, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.341878   0.35559386 0.73415124 ... 0.14034525 0.05180654 0.757296  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 5581, 8316, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.341878   0.35559386 0.73415124 ... 0.14034525 0.05180654 0.757296  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996, 17990,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.16562445 0.49762824 0.24224436 ... 0.13363037 0.         0.63426286]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[ 101, 2129, 2116, 9738, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.16562445 0.49762824 0.24224436 ... 0.13363037 0.         0.63426286]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2045,  2062,  2084,  2028, 13297,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.16562445 0.49762824 0.24224436 ... 0.13363037 0.         0.63426286]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2406,  2106,  2023, 13297, 21754,  1999,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.16562445 0.49762824 0.24224436 ... 0.13363037 0.         0.63426286]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 8582, 2194, 8617, 2023, 4946, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.16562445 0.49762824 0.24224436 ... 0.13363037 0.         0.63426286]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2406, 10836,  2023,  8582,  1005,  1055,  3954,\n",
            "         1999, 25755,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.16562445 0.49762824 0.24224436 ... 0.13363037 0.         0.63426286]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2079, 1996, 4144, 2006, 1996, 4946, 2360, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26079935 0.32836667 0.14693592 ... 0.7278496  0.36700243 0.370005  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996, 12586,  2121,  2440,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26079935 0.32836667 0.14693592 ... 0.7278496  0.36700243 0.370005  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023,  1037, 12586,  2121,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26079935 0.32836667 0.14693592 ... 0.7278496  0.36700243 0.370005  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2040, 2165, 2023, 3861, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.36505377 0.3878638  0.31764624 ... 0.23697314 0.19968995 0.35209692]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 7787, 2024, 5710, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.36505377 0.3878638  0.31764624 ... 0.23697314 0.19968995 0.35209692]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2073, 2003, 2023, 4946, 2013, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.36505377 0.3878638  0.31764624 ... 0.23697314 0.19968995 0.35209692]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 3861, 2006, 1996, 4946, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.43290675 0.08231255 0.48175588 ... 0.24215671 0.15749797 1.0264118 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996,  2711,  1005,  1055, 28095,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.43290675 0.08231255 0.48175588 ... 0.24215671 0.15749797 1.0264118 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2064, 2022, 2081, 1997, 1996, 2317, 4933, 2006, 1996,\n",
            "        2598, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.43290675 0.08231255 0.48175588 ... 0.24215671 0.15749797 1.0264118 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2003, 2016, 2725, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.36873376 0.25045305 0.40141973 ... 0.11817604 0.18496467 0.08261107]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 2023, 3696, 2047, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.36873376 0.25045305 0.40141973 ... 0.11817604 0.18496467 0.08261107]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2828, 1997, 6112, 2003, 2369, 1996, 3696, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.36873376 0.25045305 0.40141973 ... 0.11817604 0.18496467 0.08261107]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2079, 1996, 3696, 2360, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1192121  0.39355275 0.2932627  ... 0.52481383 0.39531207 0.14043272]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2158, 4147, 2006, 2010, 2132, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1192121  0.39355275 0.2932627  ... 0.52481383 0.39531207 0.14043272]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996,  2158,  3173, 14175,  6277,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1192121  0.39355275 0.2932627  ... 0.52481383 0.39531207 0.14043272]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2008, 1037, 2613, 2732, 7529, 2613, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.23357737 0.22042508 0.24365497 ... 0.32518274 0.1410828  0.08827448]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2208, 2024, 2027, 2652, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.23357737 0.22042508 0.24365497 ... 0.32518274 0.1410828  0.08827448]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 16), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996,  2158,  1999,  1996,  4281,\n",
            "         1005,  1055, 18856,  5243,  3215,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 16), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.23357737 0.22042508 0.24365497 ... 0.32518274 0.1410828  0.08827448]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2193, 2003, 2006, 1996, 2158, 1005, 1055, 2317, 9132,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.2869529  0.19304404 1.0877528  ... 0.09184217 0.11810564 0.4694744 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2003,  2016,  2652, 16568,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.2869529  0.19304404 1.0877528  ... 0.09184217 0.11810564 0.4694744 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2016, 3173, 1999, 2014, 2192, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.2869529  0.19304404 1.0877528  ... 0.09184217 0.11810564 0.4694744 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2828, 1997, 4368, 2003, 3378, 2007, 2023, 3496, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.2869529  0.19304404 1.0877528  ... 0.09184217 0.11810564 0.4694744 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2515, 2016, 2298, 3407, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.2869529  0.19304404 1.0877528  ... 0.09184217 0.11810564 0.4694744 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2051, 2003, 2009, 2006, 2014, 3422, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.2869529  0.19304404 1.0877528  ... 0.09184217 0.11810564 0.4694744 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 2715, 9982, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.98523647 0.16822103 1.1235554  ... 0.50357485 1.4252751  0.5485624 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2079, 2122, 2111, 2298, 2066, 2027, 1005, 2128, 2383, 4569,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.98523647 0.16822103 1.1235554  ... 0.50357485 1.4252751  0.5485624 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2161, 2001, 1996, 3861, 2579, 1999, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.98523647 0.16822103 1.1235554  ... 0.50357485 1.4252751  0.5485624 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 18105,  2024,  1999,  1996,  6302,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.98523647 0.16822103 1.1235554  ... 0.50357485 1.4252751  0.5485624 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 4412, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.98523647 0.16822103 1.1235554  ... 0.50357485 1.4252751  0.5485624 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 3861, 3728, 2579, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.98523647 0.16822103 1.1235554  ... 0.50357485 1.4252751  0.5485624 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 16), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2828,  1997,  3329, 16689,  2515,  1996,  2158,\n",
            "         1999,  1996,  2630,  3797,  4929,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 16), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.98523647 0.16822103 1.1235554  ... 0.50357485 1.4252751  0.5485624 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2087, 2759, 5549, 1997, 5193, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.98523647 0.16822103 1.1235554  ... 0.50357485 1.4252751  0.5485624 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 18580,  2024,  1999,  1996,  3861,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.98523647 0.16822103 1.1235554  ... 0.50357485 1.4252751  0.5485624 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[  101,  2003,  2009, 24057,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.98523647 0.16822103 1.1235554  ... 0.50357485 1.4252751  0.5485624 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2619, 2006, 2037, 3042, 2182, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.98523647 0.16822103 1.1235554  ... 0.50357485 1.4252751  0.5485624 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 2155, 2724, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.98523647 0.16822103 1.1235554  ... 0.50357485 1.4252751  0.5485624 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 2023, 1999, 2637, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.98523647 0.16822103 1.1235554  ... 0.50357485 1.4252751  0.5485624 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 2045, 2069, 2273, 1999, 1996, 6302, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.98523647 0.16822103 1.1235554  ... 0.50357485 1.4252751  0.5485624 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 3103, 5710, 1999, 2023, 3861, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.98523647 0.16822103 1.1235554  ... 0.50357485 1.4252751  0.5485624 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 2111, 2024, 1999, 1996, 3861, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73054   0.6651671 0.5511257 ... 0.5955978 1.5126507 0.5908146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 2111, 2024, 2045, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73054   0.6651671 0.5511257 ... 0.5955978 1.5126507 0.5908146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2029, 2450, 2003, 4147, 2417, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73054   0.6651671 0.5511257 ... 0.5955978 1.5126507 0.5908146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996,  3746, 14819,  2854,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73054   0.6651671 0.5511257 ... 0.5955978 1.5126507 0.5908146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1999, 1996, 2250, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73054   0.6651671 0.5511257 ... 0.5955978 1.5126507 0.5908146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 2111, 2024, 3564, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73054   0.6651671 0.5511257 ... 0.5955978 1.5126507 0.5908146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 17732,  2024,  1999,  1996,  3746,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73054   0.6651671 0.5511257 ... 0.5955978 1.5126507 0.5908146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3032, 5210, 2003, 3909, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73054   0.6651671 0.5511257 ... 0.5955978 1.5126507 0.5908146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2172, 14624,  2031,  2027,  2018,  1999,  2037,\n",
            "         3268,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73054   0.6651671 0.5511257 ... 0.5955978 1.5126507 0.5908146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2024,  2045, 22163,  1999,  1996,  3712,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73054   0.6651671 0.5511257 ... 0.5955978 1.5126507 0.5908146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 2273, 2024, 4147, 9132, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73054   0.6651671 0.5511257 ... 0.5955978 1.5126507 0.5908146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 2715, 6302, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73054   0.6651671 0.5511257 ... 0.5955978 1.5126507 0.5908146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2129, 2003, 1996, 4633, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73054   0.6651671 0.5511257 ... 0.5955978 1.5126507 0.5908146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 2679, 3586, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73054   0.6651671 0.5511257 ... 0.5955978 1.5126507 0.5908146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2029, 17678,  2024,  2122,  2111,  4637,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73054   0.6651671 0.5511257 ... 0.5955978 1.5126507 0.5908146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996,  3336, 27244,  2121,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73054   0.6651671 0.5511257 ... 0.5955978 1.5126507 0.5908146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2073, 2003, 1996, 3899, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73054   0.6651671 0.5511257 ... 0.5955978 1.5126507 0.5908146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 2045, 1037, 2843, 1997, 2111, 2012, 2023, 6891, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73054   0.6651671 0.5511257 ... 0.5955978 1.5126507 0.5908146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2450, 2725, 2007, 2014, 2398, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73054   0.6651671 0.5511257 ... 0.5955978 1.5126507 0.5908146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023,  1037, 22221,  2492,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73054   0.6651671 0.5511257 ... 0.5955978 1.5126507 0.5908146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2339, 2003, 2045, 2061, 2116, 3628, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73054   0.6651671 0.5511257 ... 0.5955978 1.5126507 0.5908146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 5568, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73054   0.6651671 0.5511257 ... 0.5955978 1.5126507 0.5908146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2024, 1996, 8397, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73054   0.6651671 0.5511257 ... 0.5955978 1.5126507 0.5908146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2009, 2055, 2000, 4040, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73054   0.6651671 0.5511257 ... 0.5955978 1.5126507 0.5908146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 2111, 2024, 1999, 2556, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73054   0.6651671 0.5511257 ... 0.5955978 1.5126507 0.5908146]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 2158, 3173, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1556949  0.28306112 0.37522566 ... 0.15699261 0.07070762 0.1378783 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2079, 2017, 2156, 7787, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1556949  0.28306112 0.37522566 ... 0.15699261 0.07070762 0.1378783 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 2111, 2024, 2045, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1556949  0.28306112 0.37522566 ... 0.15699261 0.07070762 0.1378783 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2073, 2001, 2023, 3861, 2579, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1556949  0.28306112 0.37522566 ... 0.15699261 0.07070762 0.1378783 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2073,  2003,  1996, 13528,  9935,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1556949  0.28306112 0.37522566 ... 0.15699261 0.07070762 0.1378783 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2023, 1037, 9982, 2030, 1037, 5059, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1556949  0.28306112 0.37522566 ... 0.15699261 0.07070762 0.1378783 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[ 101, 2009, 2071, 2022, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[3.6432338e-01 2.9836088e-01 1.4610237e-01 ... 8.0557925e-01 4.3874163e-02\n",
            " 3.1451634e-04]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023, 15212,  3053,  2908,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[3.6432338e-01 2.9836088e-01 1.4610237e-01 ... 8.0557925e-01 4.3874163e-02\n",
            " 3.1451634e-04]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2054,  2038,  3239, 15621,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[3.6432338e-01 2.9836088e-01 1.4610237e-01 ... 8.0557925e-01 4.3874163e-02\n",
            " 3.1451634e-04]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1999, 1996, 4281, 1997, 2023, 3861, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25374824 0.2461714  0.57371426 ... 0.30114782 0.17511289 0.15737537]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2024, 2045, 2116, 1997, 1999, 2023, 3861, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25374824 0.2461714  0.57371426 ... 0.30114782 0.17511289 0.15737537]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2129, 2116, 2024, 3756, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25374824 0.2461714  0.57371426 ... 0.30114782 0.17511289 0.15737537]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2029, 2126, 2064, 2017, 2025, 2175, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.10970713 0.23554087 0.48800015 ... 0.17786942 0.23818359 0.18888369]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[  101,  2024,  1996, 21027,  1005,  1055,  3456,  4625,  1996,\n",
            "         2168,  2126,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.10970713 0.23554087 0.48800015 ... 0.17786942 0.23818359 0.18888369]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 2158, 1005, 1055, 6007, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.10970713 0.23554087 0.48800015 ... 0.17786942 0.23818359 0.18888369]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 16), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 2447, 1005, 1055, 3125, 2000, 3113, 1996, 3608,\n",
            "        2012, 3239, 2504, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 16), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08734559 0.11731146 0.20231889 ... 0.4273069  0.1632593  0.29955825]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2515, 1996, 3902, 2031, 1016, 3798, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08734559 0.11731146 0.20231889 ... 0.4273069  0.1632593  0.29955825]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996, 20820,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08734559 0.11731146 0.20231889 ... 0.4273069  0.1632593  0.29955825]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2054,  2038,  3756, 12241,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08734559 0.11731146 0.20231889 ... 0.4273069  0.1632593  0.29955825]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2073, 2003, 1996, 3902, 4439, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08734559 0.11731146 0.20231889 ... 0.4273069  0.1632593  0.29955825]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 3345, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08734559 0.11731146 0.20231889 ... 0.4273069  0.1632593  0.29955825]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 6087, 2003, 1996, 3902, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08734559 0.11731146 0.20231889 ... 0.4273069  0.1632593  0.29955825]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 4597, 2024, 2006, 1996, 2392, 1997, 1996, 3902,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08734559 0.11731146 0.20231889 ... 0.4273069  0.1632593  0.29955825]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 7793, 2024, 2045, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08734559 0.11731146 0.20231889 ... 0.4273069  0.1632593  0.29955825]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2515,  1996,  3696,  2058,  1996,  3902,  1005,\n",
            "         1055, 19521,  2360,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.10500786 0.07133493 0.0878913  ... 0.05161627 0.5324586  0.23176637]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2339,  2024,  2048, 16825,  3760,  2084,  1996,  2060,\n",
            "         2048,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.10500786 0.07133493 0.0878913  ... 0.05161627 0.5324586  0.23176637]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 3609, 2024, 2027, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.10500786 0.07133493 0.0878913  ... 0.05161627 0.5324586  0.23176637]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2024,  2151, 16825,  1999,  1996,  2300,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.35809654 0.01488959 0.1237866  ... 0.5817987  0.7339461  0.3635403 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2515, 1996, 8351, 2298, 6530, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.35809654 0.01488959 0.1237866  ... 0.5817987  0.7339461  0.3635403 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 4111, 2956, 1999, 2023, 3861, 2019, 4639, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.35809654 0.01488959 0.1237866  ... 0.5817987  0.7339461  0.3635403 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 5568, 2665, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.35809654 0.01488959 0.1237866  ... 0.5817987  0.7339461  0.3635403 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 8351, 2024, 2045, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.35809654 0.01488959 0.1237866  ... 0.5817987  0.7339461  0.3635403 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 12559,  2024,  1999,  1996,  2492,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.2921703  0.61977875 1.1921831  ... 0.29962504 0.6182616  1.0532539 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2723, 2001, 2023, 3861, 2579, 2006, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.2921703  0.61977875 1.1921831  ... 0.29962504 0.6182616  1.0532539 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 7541, 3178, 2000, 2043, 1996, 6302, 2001,\n",
            "        2579, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.2921703  0.61977875 1.1921831  ... 0.29962504 0.6182616  1.0532539 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2073, 2003, 1996, 7668, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.1015134  0.19674112 0.13062485 ... 0.         1.6541971  0.13785434]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 3087, 1999, 2240, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.1015134  0.19674112 0.13062485 ... 0.         1.6541971  0.13785434]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 1005, 1055, 2006, 4653, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.1015134  0.19674112 0.13062485 ... 0.         1.6541971  0.13785434]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2023, 1037, 4431, 2000, 1037, 4274, 2678, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.1015134  0.19674112 0.13062485 ... 0.         1.6541971  0.13785434]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2024,  1996, 10856,  2063, 12519,  2015,  3173,  2398,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.1015134  0.19674112 0.13062485 ... 0.         1.6541971  0.13785434]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2515, 2023, 3573, 5271, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.1015134  0.19674112 0.13062485 ... 0.         1.6541971  0.13785434]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  2006,  1996, 15475,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.1015134  0.19674112 0.13062485 ... 0.         1.6541971  0.13785434]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2024,  2045, 10856,  2063, 12519,  2015,  1999,  1996,\n",
            "         6302,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.1015134  0.19674112 0.13062485 ... 0.         1.6541971  0.13785434]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 2111, 2024, 1999, 1996, 2282, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.1015134  0.19674112 0.13062485 ... 0.         1.6541971  0.13785434]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 2045, 2151, 2308, 1999, 2023, 3861, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3137657  0.36032566 0.18834655 ... 0.37373972 0.31996495 0.04484967]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 5568, 1999, 2023, 3861, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3137657  0.36032566 0.18834655 ... 0.37373972 0.31996495 0.04484967]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 1037, 4744, 2008, 3475, 1005, 1056, 3048, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3137657  0.36032566 0.18834655 ... 0.37373972 0.31996495 0.04484967]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2073, 2064, 2017, 2156, 1996, 2193, 1019, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.39497235 0.1922103  0.14778642 ... 0.76622224 0.31734318 0.03590925]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=\n",
            "array([[ 101, 2065, 2017, 2020, 1999, 2023, 2173, 2323, 2017, 4929, 1037,\n",
            "        5435, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.39497235 0.1922103  0.14778642 ... 0.76622224 0.31734318 0.03590925]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 16), dtype=int32, numpy=\n",
            "array([[ 101, 2000, 2131, 2006, 2023, 3345, 2079, 2017, 2342, 2000, 2202,\n",
            "        1037, 2770, 2707, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 16), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.39497235 0.1922103  0.14778642 ... 0.76622224 0.31734318 0.03590925]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2515, 1996, 3953, 1997, 1996, 3902, 2674, 1996, 4586, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 5418, 2003, 2006, 1996, 2723, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023, 10257,  9111,  5112,  2205,  2659,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2029, 2282, 2003, 2023, 3861, 2579, 2013, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2024,  2045, 14095,  2006,  1996,  4675,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 2813, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 1005, 1055, 7686, 1999, 1996, 5259, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2045,  1037,  7198, 28251,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2024, 1996, 3681, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2024, 1996, 8206, 5200, 2006, 1996, 2813, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 1037, 2064, 1997, 6773, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2024,  1996, 23462,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2045,  1037, 10257,  2006,  1996, 14513,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2079,  2035,  1997,  1996,  2422, 25548,  2674,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 7752, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2045,  1037, 10257,  5689,  1999,  1996,  5723,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 5259, 2005, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 24213,  2024,  5689,  2006,  1996, 14513,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996, 23462,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 23462,  2024,  2045,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2079,  2017,  2156,  1037,  3482,  1997, 14095,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2029, 3609, 2003, 7444, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 15), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2045,  1037, 11584, 18686,  3675,  2012,  1996,\n",
            "         2327,  1997,  1996,  2813,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 15), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 2422, 2006, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2024,  1996, 15220,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116,  9378, 23095,  2015,  2024, 15885,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 19), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023,  2282,  4919,  2881,  2061,  2008,  2048,\n",
            "         2111,  2064,  2079,  2048,  2367,  2477,  1998, 11834,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 19), dtype=int32, numpy=\n",
            "array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]],\n",
            "      dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2282, 2003, 2023, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 3491, 1999, 1996, 5259, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2828, 1997, 2723, 2075, 2003, 1999, 1996, 2282, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 24213,  2024,  2045,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 4597, 2024, 2045, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 1996, 3124, 1037, 9185, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2045,  1037,  3115,  2030,  2248, 13307,  5361,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2045,  1037, 10257,  2006,  1996, 10257, 14513,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01500541 0.25021926 0.2033371  ... 0.03596611 0.864492   0.21676937]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2045,  2151, 11848,  3259,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.48069784 0.21137504 0.92584157 ... 0.8403467  0.14492191 0.08136134]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 2023, 3061, 2369, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.48069784 0.21137504 0.92584157 ... 0.8403467  0.14492191 0.08136134]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2023, 2158, 3173, 1999, 2010, 2157, 2192, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.48069784 0.21137504 0.92584157 ... 0.8403467  0.14492191 0.08136134]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 5568, 1999, 1996, 6302, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.48069784 0.21137504 0.92584157 ... 0.8403467  0.14492191 0.08136134]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 2010, 6045, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.48069784 0.21137504 0.92584157 ... 0.8403467  0.14492191 0.08136134]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 2613, 2711, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.48069784 0.21137504 0.92584157 ... 0.8403467  0.14492191 0.08136134]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 10834,  2024, 15885,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3220803  0.72996426 0.30424494 ... 0.02962177 0.168987   0.00461196]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2193, 2003, 2006, 1996, 2158, 1005, 1055, 3797, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3220803  0.72996426 0.30424494 ... 0.02962177 0.168987   0.00461196]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 4368, 2003, 2108, 3491, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3220803  0.72996426 0.30424494 ... 0.02962177 0.168987   0.00461196]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2023, 2158, 2667, 2000, 2718, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.1827701  0.63013    1.345191   ... 0.         1.4604685  0.49394277]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2158, 2667, 2000, 5376, 2006, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.1827701  0.63013    1.345191   ... 0.         1.4604685  0.49394277]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 8154, 2003, 4993, 2000, 1996, 2723, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.1827701  0.63013    1.345191   ... 0.         1.4604685  0.49394277]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2045,  1037, 17260,  6277,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.45361856 0.05901484 0.4317764  ... 0.1190419  0.57196045 0.52303135]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2828, 1997, 9291, 2024, 2122, 2111, 8455, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.45361856 0.05901484 0.4317764  ... 0.1190419  0.57196045 0.52303135]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2003,  2009, 24057,  2182,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.45361856 0.05901484 0.4317764  ... 0.1190419  0.57196045 0.52303135]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2024,  1996,  4400, 13284,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.45361856 0.05901484 0.4317764  ... 0.1190419  0.57196045 0.52303135]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 16), dtype=int32, numpy=\n",
            "array([[  101,  2071,  1996,  2711,  2131,  2151,  3553,  2000,  1996,\n",
            "         2392,  1997,  1996, 14175,  6277,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 16), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.0745821  1.0724062  0.15150273 ... 0.65603244 0.99860996 0.5555912 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2023, 3295, 2087, 3497, 1037, 9201, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.0745821  1.0724062  0.15150273 ... 0.65603244 0.99860996 0.5555912 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 4111, 1999, 1996, 3746, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.0745821  1.0724062  0.15150273 ... 0.65603244 0.99860996 0.5555912 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023,  1037,  3336, 21025, 27528,  7959,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.3704296  0.0637683  0.3871818  ... 1.2441065  0.19192636 0.03435538]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2024, 29145,  2015,  2725,  1999,  1996,  3861,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.3704296  0.0637683  0.3871818  ... 1.2441065  0.19192636 0.03435538]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2024, 1996, 2187, 1016, 4176, 5983, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.3704296  0.0637683  0.3871818  ... 1.2441065  0.19192636 0.03435538]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2045, 4586, 2006, 2598, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.3704296  0.0637683  0.3871818  ... 1.2441065  0.19192636 0.03435538]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  1996, 29145,  2006,  1996,  2157,  2725,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.23682165 0.02677555 0.5118277  ... 0.63315433 0.6848166  0.3216881 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  1996,  2193,  6913,  2006,  1996, 10095,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.23682165 0.02677555 0.5118277  ... 0.63315433 0.6848166  0.3216881 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2515, 2023, 3698, 2191, 1037, 2200, 3730, 5005, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.23682165 0.02677555 0.5118277  ... 0.63315433 0.6848166  0.3216881 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 3121, 2024, 5710, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.23682165 0.02677555 0.5118277  ... 0.63315433 0.6848166  0.3216881 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 3765, 2024, 5710, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.23682165 0.02677555 0.5118277  ... 0.63315433 0.6848166  0.3216881 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2024,  2045,  2151,  8183, 13327,  2015,  2006,  1996,\n",
            "         2395,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.23682165 0.02677555 0.5118277  ... 0.63315433 0.6848166  0.3216881 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 2422, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.23682165 0.02677555 0.5118277  ... 0.63315433 0.6848166  0.3216881 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 2395, 5697, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04485437 0.13202515 0.1379375  ... 1.0270702  0.02312655 0.710569  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 2027, 2152, 1999, 1996, 2250, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04485437 0.13202515 0.1379375  ... 1.0270702  0.02312655 0.710569  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996,  2158,  4147,  1037, 10412,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04485437 0.13202515 0.1379375  ... 1.0270702  0.02312655 0.710569  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996, 21294,  2015,  3797,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04485437 0.13202515 0.1379375  ... 1.0270702  0.02312655 0.710569  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 1996, 2711, 3061, 2145, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.28046042 0.07860747 0.6455822  ... 0.4450215  0.3579389  0.3383124 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 5381, 2064, 2022, 2464, 1999, 2023, 3861, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.28046042 0.07860747 0.6455822  ... 0.4450215  0.3579389  0.3383124 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 2045, 1037, 2843, 1997, 3765, 2006, 1996, 2395, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.28046042 0.07860747 0.6455822  ... 0.4450215  0.3579389  0.3383124 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 2775, 1005, 1055, 7997, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.68414587 0.16635951 1.1609482  ... 0.00885934 0.01918713 0.28807443]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2073, 2024, 2122, 2111, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.68414587 0.16635951 1.1609482  ... 0.00885934 0.01918713 0.28807443]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 3121, 2024, 2045, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.68414587 0.16635951 1.1609482  ... 0.00885934 0.01918713 0.28807443]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2339, 2024, 1996, 2111, 3564, 2648, 2007, 5909, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.68414587 0.16635951 1.1609482  ... 0.00885934 0.01918713 0.28807443]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 5909, 2003, 2006, 1996, 2598, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.30411506 0.9601283  0.05093364 ... 0.23219737 0.3108453  0.20556992]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3074, 2003, 2023, 3746, 2013, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.30411506 0.9601283  0.05093364 ... 0.23219737 0.3108453  0.20556992]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2029, 3257, 2003, 1996, 3345, 7118, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.30411506 0.9601283  0.05093364 ... 0.23219737 0.3108453  0.20556992]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2051, 1997, 2154, 2003, 1996, 3496, 2635, 2173, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.22055916 0.10371821 0.04156941 ... 0.09025715 0.07919505 0.0459653 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=\n",
            "array([[  101,  2079,  2119,  1997,  1996,  2867,  2031,  1037,  2048,\n",
            "         2006,  2037, 11408,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.22055916 0.10371821 0.04156941 ... 0.09025715 0.07919505 0.0459653 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 2447, 1999, 1996, 2304, 6045, 3647, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.22055916 0.10371821 0.04156941 ... 0.09025715 0.07919505 0.0459653 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 7577, 2003, 2108, 2864, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.22055916 0.10371821 0.04156941 ... 0.09025715 0.07919505 0.0459653 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 4368, 2003, 2108, 2209, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.22055916 0.10371821 0.04156941 ... 0.09025715 0.07919505 0.0459653 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 2658, 2208, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.22055916 0.10371821 0.04156941 ... 0.09025715 0.07919505 0.0459653 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 5479, 3647, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.22055916 0.10371821 0.04156941 ... 0.09025715 0.07919505 0.0459653 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[ 101, 2003, 2002, 3647, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.40899274 0.7212494  0.16481987 ... 0.80376583 0.24509643 0.5779124 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2515,  1996,  2711,  2079,  2007,  2122, 16300,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.40899274 0.7212494  0.16481987 ... 0.80376583 0.24509643 0.5779124 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2515,  2023,  2711,  4929,  2019, 10792, 19688,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.40899274 0.7212494  0.16481987 ... 0.80376583 0.24509643 0.5779124 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2079, 2017, 2655, 2023, 2711, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26294833 0.2869146  0.08246481 ... 0.09732148 0.37974766 1.6765907 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2023, 1037, 3671, 2173, 2005, 1037, 4937, 2000, 3637,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26294833 0.2869146  0.08246481 ... 0.09732148 0.37974766 1.6765907 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2515, 1996, 4937, 2298, 6625, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26294833 0.2869146  0.08246481 ... 0.09732148 0.37974766 1.6765907 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 8843, 1997, 4937, 2003, 2023, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.24221411 0.4715839  0.0823318  ... 0.33310235 0.21834381 0.02281639]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 18), dtype=int32, numpy=\n",
            "array([[ 101, 2065, 1045, 2020, 4439, 1037, 2482, 2006, 2023, 2346, 2052,\n",
            "        1045, 2022, 3048, 2030, 3030, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 18), dtype=int32, numpy=\n",
            "array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]],\n",
            "      dtype=int32)>}\n",
            "[0.24221411 0.4715839  0.0823318  ... 0.33310235 0.21834381 0.02281639]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2024,  1996, 15753,  9663,  2094,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.24221411 0.4715839  0.0823318  ... 0.33310235 0.21834381 0.02281639]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 19), dtype=int32, numpy=\n",
            "array([[ 101, 2043, 2023, 4066, 1997, 4026, 8875, 2003, 2025, 3038, 2175,\n",
            "        2030, 2644, 2054, 2003, 2009, 3038, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 19), dtype=int32, numpy=\n",
            "array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]],\n",
            "      dtype=int32)>}\n",
            "[0.00899868 0.3182825  1.1915674  ... 0.41794705 0.40701532 0.15465458]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2024,  2119, 15410,  2006,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.00899868 0.3182825  1.1915674  ... 0.41794705 0.40701532 0.15465458]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 2813, 3949, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.00899868 0.3182825  1.1915674  ... 0.41794705 0.40701532 0.15465458]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2828, 1997, 3274, 2003, 2023, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.00899868 0.3182825  1.1915674  ... 0.41794705 0.40701532 0.15465458]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2003,  1996, 10437,  5507,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.00899868 0.3182825  1.1915674  ... 0.41794705 0.40701532 0.15465458]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 3746, 2006, 1996, 8080, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.00899868 0.3182825  1.1915674  ... 0.41794705 0.40701532 0.15465458]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 3274, 3898, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3669592  0.14815325 0.1155775  ... 0.42411286 0.9250572  0.3353843 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2009, 1037, 3147, 2154, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3669592  0.14815325 0.1155775  ... 0.42411286 0.9250572  0.3353843 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 16), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2009,  3497,  2023,  2711, 25817,  2055,  2893,\n",
            "         2037,  6912,  1999,  2061,  2220,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 16), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3669592  0.14815325 0.1155775  ... 0.42411286 0.9250572  0.3353843 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 3609, 1997, 1996, 2711, 1005, 1055, 9829,\n",
            "        4848, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3669592  0.14815325 0.1155775  ... 0.42411286 0.9250572  0.3353843 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 4586, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3669592  0.14815325 0.1155775  ... 0.42411286 0.9250572  0.3353843 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 16), dtype=int32, numpy=\n",
            "array([[ 101, 2052, 9132, 2022, 2019, 6413, 8875, 1997, 5929, 2005, 2023,\n",
            "        2785, 1997, 2154, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 16), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.07396873 0.13028638 0.18788756 ... 0.5901931  0.18851623 0.0392966 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2023, 1037, 2028, 2126, 2346, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.07396873 0.13028638 0.18788756 ... 0.5901931  0.18851623 0.0392966 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 2045, 3210, 2006, 1996, 2346, 2000, 3622, 4026, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.07396873 0.13028638 0.18788756 ... 0.5901931  0.18851623 0.0392966 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2828, 1997, 2482, 2003, 1996, 2665, 2482, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.05565742 0.3982104  0.22791475 ... 0.25403786 0.20947637 0.1006789 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 21025, 27528,  7959,  2015,  2024,  2045,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.05565742 0.3982104  0.22791475 ... 0.25403786 0.20947637 0.1006789 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2001,  2023,  2579,  2011,  1037,  2711,  2006, 23591,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.05565742 0.3982104  0.22791475 ... 0.25403786 0.20947637 0.1006789 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2031, 2035, 1996, 3727, 5357, 2013, 1996, 3628, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.05565742 0.3982104  0.22791475 ... 0.25403786 0.20947637 0.1006789 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2024,  1996, 21025, 27528,  7959,  2015,  2770,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.05565742 0.3982104  0.22791475 ... 0.25403786 0.20947637 0.1006789 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996, 21025, 27528,  7959, 15400,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25025734 0.14622796 0.14136776 ... 0.04447742 0.26106197 0.6281353 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 2300, 1999, 1996, 4281, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25025734 0.14622796 0.14136776 ... 0.04447742 0.26106197 0.6281353 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2024, 1996, 8351, 2725, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25025734 0.14622796 0.14136776 ... 0.04447742 0.26106197 0.6281353 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2161, 2003, 2009, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25025734 0.14622796 0.14136776 ... 0.04447742 0.26106197 0.6281353 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2828, 1997, 3899, 2003, 3666, 2058, 1996, 8351, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.23407967 0.35737374 0.39819032 ... 0.4142939  0.3029857  0.1794494 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2003,  2045,  1037, 10506,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.23407967 0.35737374 0.39819032 ... 0.4142939  0.3029857  0.1794494 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 4435, 2024, 1996, 5291, 4981, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.23407967 0.35737374 0.39819032 ... 0.4142939  0.3029857  0.1794494 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054, 14894,  2024,  1996,  5291,  4981,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.07856712 0.08993868 0.2764253  ... 0.17922527 0.01205361 0.486162  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2064, 1996, 2327, 1997, 1996, 4946, 2022, 2464, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.07856712 0.08993868 0.2764253  ... 0.17922527 0.01205361 0.486162  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2515, 2023, 4946, 4287, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.07856712 0.08993868 0.2764253  ... 0.17922527 0.01205361 0.486162  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 4946, 1999, 1996, 2250, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.07856712 0.08993868 0.2764253  ... 0.17922527 0.01205361 0.486162  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 4946, 2055, 2000, 4875, 2058, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.07856712 0.08993868 0.2764253  ... 0.17922527 0.01205361 0.486162  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 3609, 1997, 1996, 3712, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.07856712 0.08993868 0.2764253  ... 0.17922527 0.01205361 0.486162  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 4946, 4899, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.34427515 0.11295246 0.12960818 ... 0.17271915 0.70871586 0.23265287]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2040, 2003, 2023, 1037, 3861, 1997, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.34427515 0.11295246 0.12960818 ... 0.17271915 0.70871586 0.23265287]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 4368, 2003, 2023, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.34427515 0.11295246 0.12960818 ... 0.17271915 0.70871586 0.23265287]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2003, 2016, 2725, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.34427515 0.11295246 0.12960818 ... 0.17271915 0.70871586 0.23265287]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3297, 5093, 2447, 2003, 2023, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18670946 0.33094227 0.24536118 ... 0.34765422 0.8489496  1.1575276 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2024, 1996, 3756, 2477, 1999, 2037, 5551, 2005, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18670946 0.33094227 0.24536118 ... 0.34765422 0.8489496  1.1575276 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[  101,  2129,  2116, 17188,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18670946 0.33094227 0.24536118 ... 0.34765422 0.8489496  1.1575276 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 2367, 4176, 2024, 1999, 1996, 3861, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18421066 0.10492041 1.4607973  ... 0.0962334  0.50609297 0.8955211 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2003, 2023, 6994, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18421066 0.10492041 1.4607973  ... 0.0962334  0.50609297 0.8955211 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[ 101, 2054, 2003, 2751, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18421066 0.10492041 1.4607973  ... 0.0962334  0.50609297 0.8955211 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  2367,  2055,  1996,  4338,  1997,  1996,\n",
            "        16024,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.7979611  0.66464406 0.52299803 ... 0.06340584 0.20580111 0.21847369]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 2045, 2048, 2367, 4127, 1997, 4176, 2182, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.7979611  0.66464406 0.52299803 ... 0.06340584 0.20580111 0.21847369]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2024, 1996, 4176, 3061, 2279, 2000, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.7979611  0.66464406 0.52299803 ... 0.06340584 0.20580111 0.21847369]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2024,  1996, 21025, 27528,  7959,  2015,  7618,  1037,\n",
            "         3392,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.7979611  0.66464406 0.52299803 ... 0.06340584 0.20580111 0.21847369]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 29145,  2015,  2024,  1999,  2023,  3861,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.306611   0.37111723 0.7188379  ... 0.6252576  1.2095009  0.65487605]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[  101,  2003,  2009, 24057,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.306611   0.37111723 0.7188379  ... 0.6252576  1.2095009  0.65487605]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2455, 3742, 2003, 1999, 1996, 4281, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.306611   0.37111723 0.7188379  ... 0.6252576  1.2095009  0.65487605]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2009,  1037, 11559,  2154,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.51266414 0.03040188 0.21023892 ... 0.68180466 0.64611757 0.26641032]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2051, 2003, 2009, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.51266414 0.03040188 0.21023892 ... 0.68180466 0.64611757 0.26641032]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 1037, 5210, 2682, 1996, 5119, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.51266414 0.03040188 0.21023892 ... 0.68180466 0.64611757 0.26641032]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2492, 2003, 2023, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.28331226 0.31010568 0.6823931  ... 0.03796323 0.41363272 0.27559453]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 5689, 1999, 1996, 4281, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.28331226 0.31010568 0.6823931  ... 0.03796323 0.41363272 0.27559453]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 5909, 2003, 1999, 1996, 4169, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.28331226 0.31010568 0.6823931  ... 0.03796323 0.41363272 0.27559453]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[ 101, 2024, 2045, 6281, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.28331226 0.31010568 0.6823931  ... 0.03796323 0.41363272 0.27559453]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[ 101, 2129, 2116, 7251, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.28331226 0.31010568 0.6823931  ... 0.03796323 0.41363272 0.27559453]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2723, 2081, 1997, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.28331226 0.31010568 0.6823931  ... 0.03796323 0.41363272 0.27559453]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2035, 1996, 7390, 2081, 1997, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.58205587 0.5175915  0.7207092  ... 0.18016952 0.45706832 0.10710192]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2515, 1996, 2311, 2031, 1037, 4412, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.58205587 0.5175915  0.7207092  ... 0.18016952 0.45706832 0.10710192]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[  101,  2024,  1996, 21025, 27528,  7959,  2015, 12283,  2084,\n",
            "         1996,  8638,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.58205587 0.5175915  0.7207092  ... 0.18016952 0.45706832 0.10710192]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2024,  1996, 21025, 27528,  7959,  2015,  2183,  2188,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.31458256 0.23197716 1.1123815  ... 0.4263246  0.50521827 0.5402405 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2003,  2016,  4147, 27212,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.31458256 0.23197716 1.1123815  ... 0.4263246  0.50521827 0.5402405 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 2450, 1005, 1055, 2606, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.31458256 0.23197716 1.1123815  ... 0.4263246  0.50521827 0.5402405 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 2450, 5629, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.31458256 0.23197716 1.1123815  ... 0.4263246  0.50521827 0.5402405 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 5418, 1997, 1996, 2711, 1005, 1055, 3797,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.31458256 0.23197716 1.1123815  ... 0.4263246  0.50521827 0.5402405 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1999, 2392, 1997, 1996, 3203, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.31458256 0.23197716 1.1123815  ... 0.4263246  0.50521827 0.5402405 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 8088, 2559, 2041, 1996, 3332, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.31458256 0.23197716 1.1123815  ... 0.4263246  0.50521827 0.5402405 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 2450, 5962, 2000, 2204, 2739, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.31458256 0.23197716 1.1123815  ... 0.4263246  0.50521827 0.5402405 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2023, 2711, 2031, 2006, 2014, 2192, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.31458256 0.23197716 1.1123815  ... 0.4263246  0.50521827 0.5402405 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 2450, 3173, 1037, 3526, 3042, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04872409 0.6432916  0.21685162 ... 0.5621432  0.70603144 0.17410608]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 5127, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04872409 0.6432916  0.21685162 ... 0.5621432  0.70603144 0.17410608]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2515,  2009,  2272,  2007, 22201,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04872409 0.6432916  0.21685162 ... 0.5621432  0.70603144 0.17410608]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2001,  1996,  6240, 26192,  2094,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04872409 0.6432916  0.21685162 ... 0.5621432  0.70603144 0.17410608]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  9530, 21341,  2015,  2024,  2006,  1996,  5127,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04872409 0.6432916  0.21685162 ... 0.5621432  0.70603144 0.17410608]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2054,  9530, 21341,  2015,  2024,  2006,  1996,  2980,\n",
            "         3899,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04872409 0.6432916  0.21685162 ... 0.5621432  0.70603144 0.17410608]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2052, 1037, 2833, 6232, 3446, 2023, 7954, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04872409 0.6432916  0.21685162 ... 0.5621432  0.70603144 0.17410608]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[  101,  2079,  2017,  2342,  9292,  1998,  5442,  2000,  4521,\n",
            "         2023, 11642,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04872409 0.6432916  0.21685162 ... 0.5621432  0.70603144 0.17410608]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 7965, 4596, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04872409 0.6432916  0.21685162 ... 0.5621432  0.70603144 0.17410608]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  2006,  1996,  2980, 16168,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04872409 0.6432916  0.21685162 ... 0.5621432  0.70603144 0.17410608]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[ 101, 2003, 1996, 6350, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04872409 0.6432916  0.21685162 ... 0.5621432  0.70603144 0.17410608]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 7244, 1996, 9292, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04872409 0.6432916  0.21685162 ... 0.5621432  0.70603144 0.17410608]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2035, 2003, 2006, 1996, 2980, 3899, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04872409 0.6432916  0.21685162 ... 0.5621432  0.70603144 0.17410608]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 7957, 1997, 6240, 2024, 2045, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04872409 0.6432916  0.21685162 ... 0.5621432  0.70603144 0.17410608]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2052, 2017, 2404, 2006, 1996, 2833, 1999, 2023, 6302,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04872409 0.6432916  0.21685162 ... 0.5621432  0.70603144 0.17410608]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2024,  2045, 24444,  2006,  1996,  2980,  3899,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04872409 0.6432916  0.21685162 ... 0.5621432  0.70603144 0.17410608]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2515,  2016,  2156,  2014, 15642,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04872409 0.6432916  0.21685162 ... 0.5621432  0.70603144 0.17410608]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2000, 1996, 2187, 1997, 1996, 5127, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04872409 0.6432916  0.21685162 ... 0.5621432  0.70603144 0.17410608]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 6240, 2003, 2008, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04872409 0.6432916  0.21685162 ... 0.5621432  0.70603144 0.17410608]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996,  2980, 16168,  5810,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04872409 0.6432916  0.21685162 ... 0.5621432  0.70603144 0.17410608]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 2045, 2151, 8808, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04872409 0.6432916  0.21685162 ... 0.5621432  0.70603144 0.17410608]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 2980, 6077, 2024, 2045, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04872409 0.6432916  0.21685162 ... 0.5621432  0.70603144 0.17410608]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996, 20949,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11210073 0.7065018  0.04228055 ... 0.5409845  0.75056195 0.00209148]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  2006,  1996, 15168,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11210073 0.7065018  0.04228055 ... 0.5409845  0.75056195 0.00209148]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2024, 1996, 3121, 1999, 1996, 4281, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11210073 0.7065018  0.04228055 ... 0.5409845  0.75056195 0.00209148]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 3712, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1153563  0.55155325 0.74597365 ... 0.12971544 0.34630308 0.54721296]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 4589, 2015, 2024, 2045, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1153563  0.55155325 0.74597365 ... 0.12971544 0.34630308 0.54721296]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 2833, 2003, 2023, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1153563  0.55155325 0.74597365 ... 0.12971544 0.34630308 0.54721296]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2724, 2003, 2108, 6334, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1153563  0.55155325 0.74597365 ... 0.12971544 0.34630308 0.54721296]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1999, 1996, 9841, 2006, 1996, 2187, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1153563  0.55155325 0.74597365 ... 0.12971544 0.34630308 0.54721296]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2024,  2119, 15220,  3221,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1153563  0.55155325 0.74597365 ... 0.12971544 0.34630308 0.54721296]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 4605, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1153563  0.55155325 0.74597365 ... 0.12971544 0.34630308 0.54721296]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 6276, 2604, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1153563  0.55155325 0.74597365 ... 0.12971544 0.34630308 0.54721296]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2024,  2045,  2093, 15220,  1997,  2833,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1153563  0.55155325 0.74597365 ... 0.12971544 0.34630308 0.54721296]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2073, 2106, 2023, 2833, 2272, 2013, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1153563  0.55155325 0.74597365 ... 0.12971544 0.34630308 0.54721296]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 10447,  2024,  2045,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1153563  0.55155325 0.74597365 ... 0.12971544 0.34630308 0.54721296]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2785,  1997,  4111,  2003,  9831,  2075,  2006,\n",
            "         1996, 20418,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1153563  0.55155325 0.74597365 ... 0.12971544 0.34630308 0.54721296]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2024,  1996,  2048,  6090,  2015, 15885,  1996,  2168,\n",
            "         6090,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1153563  0.55155325 0.74597365 ... 0.12971544 0.34630308 0.54721296]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2828, 1997, 6240, 2003, 1999, 1996, 3861, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1153563  0.55155325 0.74597365 ... 0.12971544 0.34630308 0.54721296]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1999, 1996, 8070, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1153563  0.55155325 0.74597365 ... 0.12971544 0.34630308 0.54721296]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 5909, 2003, 2023, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1153563  0.55155325 0.74597365 ... 0.12971544 0.34630308 0.54721296]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 4589, 2477, 2008, 2003, 3013, 2039, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1153563  0.55155325 0.74597365 ... 0.12971544 0.34630308 0.54721296]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2052,  1037, 23566,  4521,  2023,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1153563  0.55155325 0.74597365 ... 0.12971544 0.34630308 0.54721296]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1999, 1996, 3585, 4605, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04604346 0.3280071  0.08401363 ... 0.34154373 0.36305314 0.2087993 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2024, 1996, 4176, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04604346 0.3280071  0.08401363 ... 0.34154373 0.36305314 0.2087993 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 5194, 2024, 2045, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04604346 0.3280071  0.08401363 ... 0.34154373 0.36305314 0.2087993 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 9540, 1996, 5194, 2024, 4815, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04604346 0.3280071  0.08401363 ... 0.34154373 0.36305314 0.2087993 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116,  2327, 16717,  2024,  2045,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.04604346 0.3280071  0.08401363 ... 0.34154373 0.36305314 0.2087993 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2785,  1997, 16358,  2003,  3491,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.24503495 0.3635285  0.04437829 ... 0.40226358 0.25897312 0.29071197]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2024, 2122, 2048, 4176, 2814, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.24503495 0.3635285  0.04437829 ... 0.40226358 0.25897312 0.29071197]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 3899, 1005, 1055, 4416, 2041, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.24503495 0.3635285  0.04437829 ... 0.40226358 0.25897312 0.29071197]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 9127, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.24503495 0.3635285  0.04437829 ... 0.40226358 0.25897312 0.29071197]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 4937, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26018614 0.33658838 0.2598673  ... 0.08803187 0.01359702 0.26168835]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 3899, 2503, 1037, 2482, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26018614 0.33658838 0.2598673  ... 0.08803187 0.01359702 0.26168835]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 3899, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26018614 0.33658838 0.2598673  ... 0.08803187 0.01359702 0.26168835]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2024, 1996, 4272, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.14714593 0.6939505  0.05354961 ... 0.6704411  0.735489   0.4198627 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2515,  2023,  4946,  2031, 23405,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.14714593 0.6939505  0.05354961 ... 0.6704411  0.735489   0.4198627 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 9738, 2024, 2045, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.14714593 0.6939505  0.05354961 ... 0.6704411  0.735489   0.4198627 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2051, 1997, 2154, 2003, 2009, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1367888  0.35600036 0.06687023 ... 0.86589974 0.6046445  0.7052945 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 6242, 2024, 1999, 1996, 2300, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1367888  0.35600036 0.06687023 ... 0.86589974 0.6046445  0.7052945 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023,  1037, 10789,  3496,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1367888  0.35600036 0.06687023 ... 0.86589974 0.6046445  0.7052945 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2024, 2122, 2111, 1999, 2300, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1638236  0.29323143 0.14253    ... 0.07857654 0.02670458 1.1001719 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2517, 2006, 1996, 4946, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1638236  0.29323143 0.14253    ... 0.07857654 0.02670458 1.1001719 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2005,  2054,  8582,  2515,  2023, 13297,  4875,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1638236  0.29323143 0.14253    ... 0.07857654 0.02670458 1.1001719 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 4916, 4946, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1638236  0.29323143 0.14253    ... 0.07857654 0.02670458 1.1001719 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 2009, 2154, 2051, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1638236  0.29323143 0.14253    ... 0.07857654 0.02670458 1.1001719 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2194, 5748, 1996, 4946, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1638236  0.29323143 0.14253    ... 0.07857654 0.02670458 1.1001719 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[  101,  2054,  4037,  2003, 17099,  2006,  1996,  2217,  1997,\n",
            "         1996,  4946,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26256108 0.67187953 0.24814904 ... 0.77748287 0.69860643 0.40885326]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  1996,  2640,  2006,  1996, 28005,  2121,\n",
            "         2000,  1996,  4373,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26256108 0.67187953 0.24814904 ... 0.77748287 0.69860643 0.40885326]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2054, 15415,  2003,  1999,  1996,  9841,  1999,  1996,\n",
            "         2067,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26256108 0.67187953 0.24814904 ... 0.77748287 0.69860643 0.40885326]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2833, 2006, 1996, 5127, 2170, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26256108 0.67187953 0.24814904 ... 0.77748287 0.69860643 0.40885326]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2024,  2045,  2151, 22688,  1999,  1996,  3861,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26256108 0.67187953 0.24814904 ... 0.77748287 0.69860643 0.40885326]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 2640, 2003, 2006, 1996, 2795, 8416, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26256108 0.67187953 0.24814904 ... 0.77748287 0.69860643 0.40885326]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116,  7766,  2024, 15885,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26256108 0.67187953 0.24814904 ... 0.77748287 0.69860643 0.40885326]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1999, 1996, 2452, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26256108 0.67187953 0.24814904 ... 0.77748287 0.69860643 0.40885326]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 2452, 2440, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26256108 0.67187953 0.24814904 ... 0.77748287 0.69860643 0.40885326]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  2665,  2006,  2122, 22094,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26256108 0.67187953 0.24814904 ... 0.77748287 0.69860643 0.40885326]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 15), dtype=int32, numpy=\n",
            "array([[  101,  2024,  2035,  1997,  1996,  9440,  3605,  1997,  1996,\n",
            "         2168,  2310, 13871,  2666,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 15), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.15320796 0.07059665 0.08061575 ... 0.09594679 0.01141369 0.13881138]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 3897, 2482, 1999, 4367, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.15320796 0.07059665 0.08061575 ... 0.09594679 0.01141369 0.13881138]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 3165, 3538, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.15320796 0.07059665 0.08061575 ... 0.09594679 0.01141369 0.13881138]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  2006,  2327,  1997,  1996,  2543, 26018,\n",
            "         3372,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.15320796 0.07059665 0.08061575 ... 0.09594679 0.01141369 0.13881138]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996, 13730,  4993,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.15320796 0.07059665 0.08061575 ... 0.09594679 0.01141369 0.13881138]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996, 26018,  3372,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.15320796 0.07059665 0.08061575 ... 0.09594679 0.01141369 0.13881138]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[ 101, 2129, 2116, 3765, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.15320796 0.07059665 0.08061575 ... 0.09594679 0.01141369 0.13881138]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 2395, 4550, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.15320796 0.07059665 0.08061575 ... 0.09594679 0.01141369 0.13881138]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 3756, 4874, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.32347867 0.4324278  0.4668337  ... 0.53234094 0.09744836 0.02980517]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2515, 1996, 3696, 2360, 2644, 2030, 3328, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.32347867 0.4324278  0.4668337  ... 0.53234094 0.09744836 0.02980517]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 2122, 2111, 3287, 2030, 2931, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.32347867 0.4324278  0.4668337  ... 0.53234094 0.09744836 0.02980517]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2024, 2122, 2111, 3173, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.1204302  0.33557492 0.9386202  ... 0.255924   0.19965608 1.708034  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 2006, 1996, 2795, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.1204302  0.33557492 0.9386202  ... 0.255924   0.19965608 1.708034  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 4111, 2003, 2279, 2000, 1996, 3203, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.1204302  0.33557492 0.9386202  ... 0.255924   0.19965608 1.708034  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 3797, 2003, 1996, 3203, 4147, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.5380113  0.20846951 0.0381314  ... 0.7605326  0.48495018 0.01658233]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2024,  2122, 29145,  2015,  2388,  1998,  2684,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.5380113  0.20846951 0.0381314  ... 0.7605326  0.48495018 0.01658233]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 2023, 1999, 3088, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.5380113  0.20846951 0.0381314  ... 0.7605326  0.48495018 0.01658233]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 4111, 2003, 1999, 1996, 6302, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.5380113  0.20846951 0.0381314  ... 0.7605326  0.48495018 0.01658233]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2024,  1996, 29145,  2015,  2725,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.5380113  0.20846951 0.0381314  ... 0.7605326  0.48495018 0.01658233]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2024, 1996, 4176, 3407, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.5380113  0.20846951 0.0381314  ... 0.7605326  0.48495018 0.01658233]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1999, 1996, 3748, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.5380113  0.20846951 0.0381314  ... 0.7605326  0.48495018 0.01658233]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2003,  1996, 29145, 15400,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09633125 0.7518339  0.37831968 ... 0.70296526 0.5361274  0.14967051]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 2879, 3173, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09633125 0.7518339  0.37831968 ... 0.70296526 0.5361274  0.14967051]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996,  2879, 20540,  2005,  1037,  6302,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09633125 0.7518339  0.37831968 ... 0.70296526 0.5361274  0.14967051]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 2023, 2775, 1005, 1055, 3797, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09633125 0.7518339  0.37831968 ... 0.70296526 0.5361274  0.14967051]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 1037, 3417, 8962, 3723, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09633125 0.7518339  0.37831968 ... 0.70296526 0.5361274  0.14967051]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2052, 2023, 2011, 2066, 2000, 2718, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09633125 0.7518339  0.37831968 ... 0.70296526 0.5361274  0.14967051]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 2879, 3201, 2005, 1996, 2350, 8121, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09633125 0.7518339  0.37831968 ... 0.70296526 0.5361274  0.14967051]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2023, 2711, 4147, 2006, 2037, 2132, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09633125 0.7518339  0.37831968 ... 0.70296526 0.5361274  0.14967051]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2023, 2711, 1999, 1037, 3598, 6375, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09633125 0.7518339  0.37831968 ... 0.70296526 0.5361274  0.14967051]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 2879, 3797, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09633125 0.7518339  0.37831968 ... 0.70296526 0.5361274  0.14967051]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 15), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 2398, 2003, 1996, 2879, 2478, 2000, 2907, 1996,\n",
            "        3598, 7151, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 15), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09633125 0.7518339  0.37831968 ... 0.70296526 0.5361274  0.14967051]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 2336, 2024, 1999, 1996, 6302, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09417908 0.38917297 0.1627704  ... 0.5248115  0.08194818 0.39170623]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2515,  2002,  2298, 24305,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09417908 0.38917297 0.1627704  ... 0.5248115  0.08194818 0.39170623]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2095, 2001, 2023, 6302, 2579, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09417908 0.38917297 0.1627704  ... 0.5248115  0.08194818 0.39170623]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2350, 7419, 2724, 2003, 2002, 1037, 2112, 1997, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09417908 0.38917297 0.1627704  ... 0.5248115  0.08194818 0.39170623]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2193, 2003, 1996, 2158, 4147, 2006, 2010, 3797, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.13431612 0.45311743 0.11884706 ... 0.31157506 0.5057813  0.16815755]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116,  2123, 16446,  2024,  2045,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.13431612 0.45311743 0.11884706 ... 0.31157506 0.5057813  0.16815755]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 4338, 2003, 1996, 2123, 4904, 2006, 1996, 2327, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.13431612 0.45311743 0.11884706 ... 0.31157506 0.5057813  0.16815755]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  1996, 14902,  2006,  1996,  2123,  4904,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.66617    0.57178116 0.25076738 ... 0.53640103 0.31200507 0.01352468]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 2111, 2024, 2464, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.66617    0.57178116 0.25076738 ... 0.53640103 0.31200507 0.01352468]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 4633, 2066, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.66617    0.57178116 0.25076738 ... 0.53640103 0.31200507 0.01352468]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 2613, 2775, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.66617    0.57178116 0.25076738 ... 0.53640103 0.31200507 0.01352468]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 2019, 3746, 2006, 1996, 2217, 1997, 1996, 3345,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.66617    0.57178116 0.25076738 ... 0.53640103 0.31200507 0.01352468]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3430, 2001, 2109, 2000, 4339, 2006, 1996, 3696, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.66617    0.57178116 0.25076738 ... 0.53640103 0.31200507 0.01352468]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 4132, 2024, 1996, 2111, 2006, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3483976  0.23374969 0.5688051  ... 0.64899886 0.77183264 0.0178744 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 4338, 2003, 1996, 2644, 3696, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3483976  0.23374969 0.5688051  ... 0.64899886 0.77183264 0.0178744 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2073, 2003, 1996, 3696, 7458, 2017, 2013, 2183, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3483976  0.23374969 0.5688051  ... 0.64899886 0.77183264 0.0178744 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2515, 1996, 2422, 2079, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1542947  0.40875882 0.46020508 ... 0.19795376 0.14234565 0.8710488 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2828, 1997, 5210, 2003, 2008, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1542947  0.40875882 0.46020508 ... 0.19795376 0.14234565 0.8710488 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 5472, 1999, 1996, 3746, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1542947  0.40875882 0.46020508 ... 0.19795376 0.14234565 0.8710488 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2029,  2126,  2003,  1996,  3612, 11221,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.03669008 0.23823683 0.477466   ... 0.47849914 0.26180625 0.02222327]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3632, 1999, 2122, 6470, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.03669008 0.23823683 0.477466   ... 0.47849914 0.26180625 0.02222327]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2024,  1996, 18781,  2015, 14231,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.03669008 0.23823683 0.477466   ... 0.47849914 0.26180625 0.02222327]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2024,  2122,  6470, 13338,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.03669008 0.23823683 0.477466   ... 0.47849914 0.26180625 0.02222327]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2828,  1997,  2813,  2003,  2369,  1996, 18781,\n",
            "         2015,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.03669008 0.23823683 0.477466   ... 0.47849914 0.26180625 0.02222327]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 2023, 4874, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01771626 0.0177247  1.7826715  ... 0.5430994  0.20411274 0.07856432]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2003,  1996, 18781,  2715,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01771626 0.0177247  1.7826715  ... 0.5430994  0.20411274 0.07856432]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 4870, 2024, 2122, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.01771626 0.0177247  1.7826715  ... 0.5430994  0.20411274 0.07856432]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  5418,  2003,  2006,  1996, 18781,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5771637  0.09195909 0.48558637 ... 0.86336327 0.7164309  0.47347167]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2103, 2003, 2023, 5119, 3578, 1999, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5771637  0.09195909 0.48558637 ... 0.86336327 0.7164309  0.47347167]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2024, 1996, 2751, 5200, 2006, 1996, 5119, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5771637  0.09195909 0.48558637 ... 0.86336327 0.7164309  0.47347167]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2079, 2119, 3903, 1997, 1996, 5119, 2674, 2335, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5771637  0.09195909 0.48558637 ... 0.86336327 0.7164309  0.47347167]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2008, 1037, 2235, 5119, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3461661  0.447603   0.16120878 ... 0.5613526  0.09285765 0.48531646]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 7829, 2194, 2003, 3421, 1999, 1996, 3496, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3461661  0.447603   0.16120878 ... 0.5613526  0.09285765 0.48531646]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 2111, 2024, 4147, 9132, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3461661  0.447603   0.16120878 ... 0.5613526  0.09285765 0.48531646]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[ 101, 2129, 2116, 2111, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3461661  0.447603   0.16120878 ... 0.5613526  0.09285765 0.48531646]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2024, 1996, 2111, 3061, 2006, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3461661  0.447603   0.16120878 ... 0.5613526  0.09285765 0.48531646]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 2151, 1997, 1996, 2111, 1999, 4367, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.39554352 0.27906513 0.27491164 ... 0.80414504 0.72500634 0.52526104]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 4111, 3491, 2619, 1005, 1055, 9004, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.39554352 0.27906513 0.27491164 ... 0.80414504 0.72500634 0.52526104]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2003,  2023,  6302, 14231,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.39554352 0.27906513 0.27491164 ... 0.80414504 0.72500634 0.52526104]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2339,  2024,  1996, 17977,  2006,  1996,  5568,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.39554352 0.27906513 0.27491164 ... 0.80414504 0.72500634 0.52526104]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 8351, 3647, 1999, 2023, 2181, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26904327 1.1307429  0.01351812 ... 0.00253845 0.84761274 0.81899357]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 2122, 4620, 1997, 1996, 2168, 6077, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26904327 1.1307429  0.01351812 ... 0.00253845 0.84761274 0.81899357]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2079, 1996, 6077, 2298, 7965, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26904327 1.1307429  0.01351812 ... 0.00253845 0.84761274 0.81899357]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 4620, 2024, 2045, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26488525 0.21469036 0.10717871 ... 0.12031239 0.136188   0.13147354]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2515, 1996, 5119, 2298, 2715, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26488525 0.21469036 0.10717871 ... 0.12031239 0.136188   0.13147354]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 1000, 3347, 1000, 2003, 3805, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26488525 0.21469036 0.10717871 ... 0.12031239 0.136188   0.13147354]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2051, 2003, 2009, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.05861124 0.06496678 0.41364467 ... 0.7955303  0.07676027 0.32409936]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 6045, 2003, 1996, 2158, 2006, 1996, 4049,\n",
            "        4147, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.05861124 0.06496678 0.41364467 ... 0.7955303  0.07676027 0.32409936]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2158, 2006, 1996, 4049, 2725, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.05861124 0.06496678 0.41364467 ... 0.7955303  0.07676027 0.32409936]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 5055, 2024, 1999, 1996, 3861, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06994434 0.02537311 0.615668   ... 0.30287957 0.16550408 1.1750408 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2640,  2003,  2006,  1996, 10257,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06994434 0.02537311 0.615668   ... 0.30287957 0.16550408 1.1750408 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2045,  1037, 18676,  3886, 10439, 15204,  3401,\n",
            "         1999,  1996,  3861,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06994434 0.02537311 0.615668   ... 0.30287957 0.16550408 1.1750408 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2031,  2027,  7429,  1996, 18097,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06994434 0.02537311 0.615668   ... 0.30287957 0.16550408 1.1750408 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2003,  1996, 17428,  2006,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06994434 0.02537311 0.615668   ... 0.30287957 0.16550408 1.1750408 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996, 10257,  2000,  1996,  2187,  4638,  6850,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06994434 0.02537311 0.615668   ... 0.30287957 0.16550408 1.1750408 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  4204,  1996, 16247,  2327,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06994434 0.02537311 0.615668   ... 0.30287957 0.16550408 1.1750408 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2024,  1996, 24213,  6530,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06994434 0.02537311 0.615668   ... 0.30287957 0.16550408 1.1750408 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2024,  1996, 16247,  1998, 17428,  2028, 10439, 15204,\n",
            "         3401,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06994434 0.02537311 0.615668   ... 0.30287957 0.16550408 1.1750408 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  1999,  1996, 11015,  2006,  1996,  4675,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.06994434 0.02537311 0.615668   ... 0.30287957 0.16550408 1.1750408 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 5239, 2330, 2030, 2701, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.02962514 0.6761943  0.09192066 ... 0.22919558 0.07239352 0.10081445]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 4499, 2024, 1999, 2006, 1996, 3162, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.02962514 0.6761943  0.09192066 ... 0.22919558 0.07239352 0.10081445]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2103, 2003, 2023, 2284, 1999, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.02962514 0.6761943  0.09192066 ... 0.22919558 0.07239352 0.10081445]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 1996, 4499, 2183, 1999, 1996, 2168, 3257, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.02962514 0.6761943  0.09192066 ... 0.22919558 0.07239352 0.10081445]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2373, 3120, 2079, 1996, 4499, 3491, 2224, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.07863943 0.44524506 0.39935705 ... 1.1727273  0.17458543 0.2611068 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 4845, 5870, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.07863943 0.44524506 0.39935705 ... 1.1727273  0.17458543 0.2611068 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 16), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2023, 3861, 2579, 1010, 2763, 1010, 1999, 1037, 2188,\n",
            "        2030, 1037, 4825, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 16), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.07863943 0.44524506 0.39935705 ... 1.1727273  0.17458543 0.2611068 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 2711, 1999, 1996, 6302, 6886, 2833, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.07863943 0.44524506 0.39935705 ... 1.1727273  0.17458543 0.2611068 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2040, 2003, 4147, 7877, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.07863943 0.44524506 0.39935705 ... 1.1727273  0.17458543 0.2611068 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2003,  1996,  2795, 18307,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.07863943 0.44524506 0.39935705 ... 1.1727273  0.17458543 0.2611068 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2515, 1996, 2775, 2298, 3407, 2030, 6517, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.07204325 0.5533635  0.06589664 ... 0.18675688 0.07454331 0.08329955]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 3899, 6680, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.07204325 0.5533635  0.06589664 ... 0.18675688 0.07454331 0.08329955]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2029, 2126, 2003, 1996, 3899, 1005, 1055, 2187, 4540, 7302,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.07204325 0.5533635  0.06589664 ... 0.18675688 0.07454331 0.08329955]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  6087,  2024,  1996, 10005,  2369,  1996,  3899,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.17992061 0.15741402 0.08324455 ... 0.5467254  0.6015876  0.26618913]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[  101,  2003,  2023, 12701,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.17992061 0.15741402 0.08324455 ... 0.5467254  0.6015876  0.26618913]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996,  2158,  4147, 13178,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.17992061 0.15741402 0.08324455 ... 0.5467254  0.6015876  0.26618913]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2038, 2023, 2711, 5357, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.17992061 0.15741402 0.08324455 ... 0.5467254  0.6015876  0.26618913]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2158, 4147, 2006, 2010, 2067, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.17992061 0.15741402 0.08324455 ... 0.5467254  0.6015876  0.26618913]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 3712, 3154, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.17992061 0.15741402 0.08324455 ... 0.5467254  0.6015876  0.26618913]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 2111, 2024, 1999, 1996, 3861, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.17992061 0.15741402 0.08324455 ... 0.5467254  0.6015876  0.26618913]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2339,  2003,  1996,  3061,  2158,  4147, 17072,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.17992061 0.15741402 0.08324455 ... 0.5467254  0.6015876  0.26618913]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 1996, 2158, 3048, 2855, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.17992061 0.15741402 0.08324455 ... 0.5467254  0.6015876  0.26618913]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2515, 2023, 2158, 4929, 1037, 7415, 6598, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.17992061 0.15741402 0.08324455 ... 0.5467254  0.6015876  0.26618913]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 2045, 3628, 1999, 2023, 6302, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.17992061 0.15741402 0.08324455 ... 0.5467254  0.6015876  0.26618913]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2515, 1996, 2711, 2031, 2006, 2010, 2132, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.17992061 0.15741402 0.08324455 ... 0.5467254  0.6015876  0.26618913]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2024,  1996, 10567,  7244,  1996,  4586,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.17992061 0.15741402 0.08324455 ... 0.5467254  0.6015876  0.26618913]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2038,  2045,  2042,  2060,  2111, 12701,  2182,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.17992061 0.15741402 0.08324455 ... 0.5467254  0.6015876  0.26618913]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 2711, 2725, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.17992061 0.15741402 0.08324455 ... 0.5467254  0.6015876  0.26618913]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2023, 2711, 4147, 1037, 6045, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1497356  0.14018725 0.3828455  ... 0.18595177 0.08121375 0.        ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 4743, 1999, 3462, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1497356  0.14018725 0.3828455  ... 0.18595177 0.08121375 0.        ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2024,  1996,  4176,  2956,  1999,  1996,  3861, 11993,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1497356  0.14018725 0.3828455  ... 0.18595177 0.08121375 0.        ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2024, 1996, 5055, 2304, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1497356  0.14018725 0.3828455  ... 0.18595177 0.08121375 0.        ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2079, 2017, 2228, 2023, 4743, 2453, 2022, 2559, 2005,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1497356  0.14018725 0.3828455  ... 0.18595177 0.08121375 0.        ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 4111, 2003, 1999, 1996, 5568, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1497356  0.14018725 0.3828455  ... 0.18595177 0.08121375 0.        ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 4743, 2725, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1497356  0.14018725 0.3828455  ... 0.18595177 0.08121375 0.        ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2003,  2023,  4743, 23990,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1497356  0.14018725 0.3828455  ... 0.18595177 0.08121375 0.        ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 2045, 2062, 2084, 1016, 5055, 1999, 2023, 3861, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1497356  0.14018725 0.3828455  ... 0.18595177 0.08121375 0.        ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2079, 2122, 5055, 2444, 2379, 2300, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1497356  0.14018725 0.3828455  ... 0.18595177 0.08121375 0.        ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[ 101, 2003, 2045, 2300, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1497356  0.14018725 0.3828455  ... 0.18595177 0.08121375 0.        ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 4281, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1497356  0.14018725 0.3828455  ... 0.18595177 0.08121375 0.        ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 1996, 4111, 1999, 2929, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11142467 0.09850298 0.24541198 ... 0.07671246 0.49369624 0.27029538]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2024, 1996, 3057, 3173, 1999, 2037, 2398, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11142467 0.09850298 0.24541198 ... 0.07671246 0.49369624 0.27029538]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 1996, 2111, 3403, 2000, 2892, 1996, 2395, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11142467 0.09850298 0.24541198 ... 0.07671246 0.49369624 0.27029538]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 4744, 2517, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11142467 0.09850298 0.24541198 ... 0.07671246 0.49369624 0.27029538]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2828, 1997, 4316, 2003, 2023, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11142467 0.09850298 0.24541198 ... 0.07671246 0.49369624 0.27029538]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 16), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  1996,  2034,  3661,  2006,  1996,  6105,\n",
            "         5127,  1997,  1996, 15805,  4744,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 16), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11142467 0.09850298 0.24541198 ... 0.07671246 0.49369624 0.27029538]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2038, 1996, 4744, 2042, 2517, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.33605438 0.8097403  0.06714444 ... 0.110466   0.34893247 0.60132366]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2023, 4111, 4141, 2179, 1999, 9201, 2015, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.33605438 0.8097403  0.06714444 ... 0.110466   0.34893247 0.60132366]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 2014, 4377, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.33605438 0.8097403  0.06714444 ... 0.110466   0.34893247 0.60132366]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 6468, 2024, 2045, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.33605438 0.8097403  0.06714444 ... 0.110466   0.34893247 0.60132366]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 4562, 3061, 2006, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.33605438 0.8097403  0.06714444 ... 0.110466   0.34893247 0.60132366]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996,  4937,  3173,  1996, 10658,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.12629242 0.04844256 0.38441175 ... 0.18464884 0.08969273 0.15501589]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2071,  2022,  1999,  1996, 26536,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.12629242 0.04844256 0.38441175 ... 0.18464884 0.08969273 0.15501589]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2515, 1996, 2158, 2031, 1037, 3797, 2006, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.12629242 0.04844256 0.38441175 ... 0.18464884 0.08969273 0.15501589]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2515, 2023, 2711, 2298, 2205, 4010, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3534028  0.28488684 0.99902475 ... 0.72411597 0.45112386 1.030729  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996, 11851,  2008,  1996, 10447,\n",
            "         2024,  2006,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3534028  0.28488684 0.99902475 ... 0.72411597 0.45112386 1.030729  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116,  7766,  2006,  1996, 11851,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3534028  0.28488684 0.99902475 ... 0.72411597 0.45112386 1.030729  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996,  7759, 16283,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3534028  0.28488684 0.99902475 ... 0.72411597 0.45112386 1.030729  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 3087, 2635, 1996, 7954, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3534028  0.28488684 0.99902475 ... 0.72411597 0.45112386 1.030729  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2417, 2008, 2003, 2872, 2006, 1996, 5127, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3534028  0.28488684 0.99902475 ... 0.72411597 0.45112386 1.030729  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2828,  1997, 10268,  2024,  2006,  1996, 11851,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3534028  0.28488684 0.99902475 ... 0.72411597 0.45112386 1.030729  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 10447,  2024,  2045,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6253369  0.17353007 0.44258443 ... 0.10846957 0.11609883 0.5565529 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 2111, 2024, 3061, 2011, 1996, 8638, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6253369  0.17353007 0.44258443 ... 0.10846957 0.11609883 0.5565529 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 17977,  2024, 15400,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6253369  0.17353007 0.44258443 ... 0.10846957 0.11609883 0.5565529 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 3899, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6253369  0.17353007 0.44258443 ... 0.10846957 0.11609883 0.5565529 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2009, 11559,  2030, 24706,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6253369  0.17353007 0.44258443 ... 0.10846957 0.11609883 0.5565529 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 1037, 2160, 2369, 1996, 3628, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6253369  0.17353007 0.44258443 ... 0.10846957 0.11609883 0.5565529 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2024, 1996, 8044, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6253369  0.17353007 0.44258443 ... 0.10846957 0.11609883 0.5565529 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 4176, 2024, 2216, 1999, 1996, 5568, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6253369  0.17353007 0.44258443 ... 0.10846957 0.11609883 0.5565529 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2024, 1996, 4206, 3628, 2170, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.153771   0.19335319 0.0849853  ... 0.3808047  0.7036916  0.23248199]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 2598, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.153771   0.19335319 0.0849853  ... 0.3808047  0.7036916  0.23248199]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 2010, 6045, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.153771   0.19335319 0.0849853  ... 0.3808047  0.7036916  0.23248199]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 2711, 2725, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.153771   0.19335319 0.0849853  ... 0.3808047  0.7036916  0.23248199]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023, 21294,  3048,  3435,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.153771   0.19335319 0.0849853  ... 0.3808047  0.7036916  0.23248199]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2711, 3173, 1999, 2010, 2398, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.382943   0.30464387 0.14133802 ... 0.6569194  0.13814883 0.07781204]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 4004, 2406, 2003, 2023, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.382943   0.30464387 0.14133802 ... 0.6569194  0.13814883 0.07781204]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2006, 1996, 2187, 2217, 1997, 1996, 3345, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.382943   0.30464387 0.14133802 ... 0.6569194  0.13814883 0.07781204]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2193, 2003, 2006, 1996, 6536, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.382943   0.30464387 0.14133802 ... 0.6569194  0.13814883 0.07781204]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 4499, 2024, 2045, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.03465001 0.10185748 0.52993953 ... 0.11581872 0.46718323 0.40519744]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 6773, 2003, 2006, 1996, 2067, 2813, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.03465001 0.10185748 0.52993953 ... 0.11581872 0.46718323 0.40519744]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 5689, 2006, 1996, 2813, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.03465001 0.10185748 0.52993953 ... 0.11581872 0.46718323 0.40519744]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996, 20452,  2006,  1996,  2723,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.22635454 0.26937962 0.13182984 ... 0.394257   0.6514583  0.52863276]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 2813, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.22635454 0.26937962 0.13182984 ... 0.394257   0.6514583  0.52863276]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2282, 2003, 2008, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.22635454 0.26937962 0.13182984 ... 0.394257   0.6514583  0.52863276]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2024, 2045, 5108, 2183, 2039, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.22635454 0.26937962 0.13182984 ... 0.394257   0.6514583  0.52863276]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2282, 2003, 2023, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.22635454 0.26937962 0.13182984 ... 0.394257   0.6514583  0.52863276]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2003,  1996,  6302, 14231,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.22635454 0.26937962 0.13182984 ... 0.394257   0.6514583  0.52863276]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 2019, 2041, 4580, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.22635454 0.26937962 0.13182984 ... 0.394257   0.6514583  0.52863276]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023,  1996,  2168,  3861,  2030,  1037,  8902,\n",
            "        20679,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.22635454 0.26937962 0.13182984 ... 0.394257   0.6514583  0.52863276]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 2311, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.22635454 0.26937962 0.13182984 ... 0.394257   0.6514583  0.52863276]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2073,  2003,  1996,  4799, 11848,  2835,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.908813   0.37674856 0.7225749  ... 0.01281686 0.5498296  1.2915711 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2108, 2853, 1999, 1996, 3006, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.908813   0.37674856 0.7225749  ... 0.01281686 0.5498296  1.2915711 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 2833, 2003, 2006, 1996, 2521, 3953, 2187,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.908813   0.37674856 0.7225749  ... 0.01281686 0.5498296  1.2915711 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2073, 2003, 2023, 3861, 2635, 2173, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.908813   0.37674856 0.7225749  ... 0.01281686 0.5498296  1.2915711 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2029, 10962,  2024,  2122,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.908813   0.37674856 0.7225749  ... 0.01281686 0.5498296  1.2915711 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 2311, 2064, 2022, 2464, 1999, 1996, 4281,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.908813   0.37674856 0.7225749  ... 0.01281686 0.5498296  1.2915711 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3257, 2003, 1996, 2158, 2559, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.02564645 0.08290698 0.11421417 ... 0.27195013 0.903779   0.06798064]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996, 14175,  6277,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.02564645 0.08290698 0.11421417 ... 0.27195013 0.903779   0.06798064]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2003,  1996,  2450, 19967,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.02564645 0.08290698 0.11421417 ... 0.27195013 0.903779   0.06798064]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 2450, 1005, 1055, 2606, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.10718906 0.08826548 0.10331762 ... 0.05664675 0.0336451  0.8225323 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 2045, 8044, 1999, 1996, 3712, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.10718906 0.08826548 0.10331762 ... 0.05664675 0.0336451  0.8225323 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 4144, 2024, 2006, 1996, 5725, 1997, 1996, 4946, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.10718906 0.08826548 0.10331762 ... 0.05664675 0.0336451  0.8225323 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2597, 1997, 1996, 4899, 6718, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.10718906 0.08826548 0.10331762 ... 0.05664675 0.0336451  0.8225323 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2194, 2515, 2023, 4946, 7141, 2000, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.10718906 0.08826548 0.10331762 ... 0.05664675 0.0336451  0.8225323 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 5725, 1997, 1996, 4946, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.10718906 0.08826548 0.10331762 ... 0.05664675 0.0336451  0.8225323 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054, 10029,  1999,  1996,  3712,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25670666 0.35774362 1.1228547  ... 0.5983412  0.5894692  0.13758841]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2785,  1997, 11546,  2024,  3491,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25670666 0.35774362 1.1228547  ... 0.5983412  0.5894692  0.13758841]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2515, 2023, 2298, 2000, 2031, 2172, 5699, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25670666 0.35774362 1.1228547  ... 0.5983412  0.5894692  0.13758841]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2038,  2042,  3013,  2046, 25609,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25670666 0.35774362 1.1228547  ... 0.5983412  0.5894692  0.13758841]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 10733,  2015,  2024,  2045,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25670666 0.35774362 1.1228547  ... 0.5983412  0.5894692  0.13758841]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996,  2795, 23095,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25670666 0.35774362 1.1228547  ... 0.5983412  0.5894692  0.13758841]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 8785, 2003, 2023, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.43395573 0.12401149 0.86503804 ... 0.17733124 0.3852289  0.22520275]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 15), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 5080, 2006, 1996, 2187, 1997, 1996, 3861,\n",
            "        2109, 2005, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 15), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.43395573 0.12401149 0.86503804 ... 0.17733124 0.3852289  0.22520275]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2691, 3609, 1997, 1996, 2093, 5733, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.43395573 0.12401149 0.86503804 ... 0.17733124 0.3852289  0.22520275]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 5080, 2012, 1996, 2327, 1997, 1996, 3861,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.9482559  0.4089952  1.7298328  ... 0.08907208 0.18440866 0.66434765]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2024, 1996, 5751, 1999, 2394, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.9482559  0.4089952  1.7298328  ... 0.08907208 0.18440866 0.66434765]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2171, 1997, 2023, 2395, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.9482559  0.4089952  1.7298328  ... 0.08907208 0.18440866 0.66434765]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2073,  2024,  1996, 26191,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.9482559  0.4089952  1.7298328  ... 0.08907208 0.18440866 0.66434765]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2073, 2001, 2023, 3861, 2579, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.9482559  0.4089952  1.7298328  ... 0.08907208 0.18440866 0.66434765]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2171, 1997, 1996, 2082, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.9482559  0.4089952  1.7298328  ... 0.08907208 0.18440866 0.66434765]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 16), dtype=int32, numpy=\n",
            "array([[  101,  2054, 12760,  2175,  2046,  1996,  4392,  2008,  2003,\n",
            "         3205,  2117,  2013,  1996,  3953,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 16), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.8875065  0.14977261 0.12005116 ... 0.6123308  0.2453945  0.4429258 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1999, 1037, 2697, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.8875065  0.14977261 0.12005116 ... 0.6123308  0.2453945  0.4429258 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2023, 7820, 2604, 2109, 2005, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.8875065  0.14977261 0.12005116 ... 0.6123308  0.2453945  0.4429258 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2002, 19967,  2157,  2085,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.23442268 0.09671588 0.14259504 ... 0.11422104 0.0103723  0.12053729]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2208, 2003, 2108, 2209, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.23442268 0.09671588 0.14259504 ... 0.11422104 0.0103723  0.12053729]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 6007, 2024, 2027, 4147, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.23442268 0.09671588 0.14259504 ... 0.11422104 0.0103723  0.12053729]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 3608, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.4512429  0.3907897  0.21302378 ... 0.17352799 0.7948184  0.32065892]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 3899, 2652, 2007, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.4512429  0.3907897  0.21302378 ... 0.17352799 0.7948184  0.32065892]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 3899, 2003, 2023, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.4512429  0.3907897  0.21302378 ... 0.17352799 0.7948184  0.32065892]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2073, 2003, 2023, 2380, 2284, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3560009  0.15222146 0.62174964 ... 0.6133013  0.14974025 0.02277461]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 2346, 4742, 2003, 1996, 2879, 3061, 2104,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3560009  0.15222146 0.62174964 ... 0.6133013  0.14974025 0.02277461]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2323, 2023, 2346, 3696, 2022, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3560009  0.15222146 0.62174964 ... 0.6133013  0.14974025 0.02277461]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2073, 2003, 1996, 2879, 2559, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20642586 0.2733751  0.07260498 ... 0.03542002 0.22705087 0.15705411]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2013,  2029,  4796,  2097,  1996,  5467,  2022, 15971,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20642586 0.2733751  0.07260498 ... 0.03542002 0.22705087 0.15705411]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2193, 2003, 2006, 1996, 3696, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20642586 0.2733751  0.07260498 ... 0.03542002 0.22705087 0.15705411]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996, 13297,  1037,  2309,  1011,  3194,  4946,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20642586 0.2733751  0.07260498 ... 0.03542002 0.22705087 0.15705411]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2024,  1996, 13560,  2006,  1996,  4946,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20642586 0.2733751  0.07260498 ... 0.03542002 0.22705087 0.15705411]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 5324, 2003, 1996, 5581, 7381, 1999, 1996, 2067,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09015125 0.09607668 0.23866871 ... 0.63451636 0.27315712 0.03832169]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 4286, 2024, 1999, 1996, 6302, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09015125 0.09607668 0.23866871 ... 0.63451636 0.27315712 0.03832169]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3661,  2003,  2006,  1996,  5093, 14513,  3388,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09015125 0.09607668 0.23866871 ... 0.63451636 0.27315712 0.03832169]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2024, 1996, 2447, 1005, 1055, 5093, 6007, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09015125 0.09607668 0.23866871 ... 0.63451636 0.27315712 0.03832169]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 2447, 1999, 3558, 3967, 2007, 1996, 2457, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09015125 0.09607668 0.23866871 ... 0.63451636 0.27315712 0.03832169]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 17), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 3609, 1997, 1996, 6045, 1996, 2158, 2003,\n",
            "        4147, 2013, 1996, 4281, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 17), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09015125 0.09607668 0.23866871 ... 0.63451636 0.27315712 0.03832169]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 15), dtype=int32, numpy=\n",
            "array([[ 101, 2029, 1997, 1996, 2158, 1005, 1055, 2519, 2024, 4257, 2006,\n",
            "        1996, 2598, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 15), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09015125 0.09607668 0.23866871 ... 0.63451636 0.27315712 0.03832169]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2208, 2003, 2002, 2652, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3261646  0.27901447 0.35385162 ... 0.00929355 0.13202971 0.26207522]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2024,  1996,  4589, 23825,  2005,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3261646  0.27901447 0.35385162 ... 0.00929355 0.13202971 0.26207522]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 2023, 2005, 9045, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3261646  0.27901447 0.35385162 ... 0.00929355 0.13202971 0.26207522]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 15), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2515,  1996, 13171,  2006,  2327,  1997,  1996,\n",
            "         2543, 26018,  3372,  5769,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 15), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3261646  0.27901447 0.35385162 ... 0.00929355 0.13202971 0.26207522]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996, 13171,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3261646  0.27901447 0.35385162 ... 0.00929355 0.13202971 0.26207522]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996,  2300, 18054,  2041,  2046,  1996,  2395,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3261646  0.27901447 0.35385162 ... 0.00929355 0.13202971 0.26207522]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996,  2543, 26018,  3372, 14309,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3261646  0.27901447 0.35385162 ... 0.00929355 0.13202971 0.26207522]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 2395, 3420, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73423696 0.14262006 0.29296803 ... 0.20395133 0.33439937 0.09914523]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 15), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3019, 2486, 2003, 7645, 2087, 4415, 1999, 2023, 2186,\n",
            "        1997, 4620, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 15), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73423696 0.14262006 0.29296803 ... 0.20395133 0.33439937 0.09914523]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2052,  2017,  2113,  2065,  2045,  2003,  1037,\n",
            "        10798,  3518,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73423696 0.14262006 0.29296803 ... 0.20395133 0.33439937 0.09914523]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  1996,  2171,  1997,  1996, 10798,  2276,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.73423696 0.14262006 0.29296803 ... 0.20395133 0.33439937 0.09914523]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  1996,  2193,  1997,  1996,  3313, 20946,\n",
            "         3902,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09776252 0.14000922 0.17981765 ... 0.38904458 0.17996795 0.3174233 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[  101,  2003,  2009, 11559,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09776252 0.14000922 0.17981765 ... 0.38904458 0.17996795 0.3174233 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 5344, 2052, 2023, 5119, 2031, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09776252 0.14000922 0.17981765 ... 0.38904458 0.17996795 0.3174233 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 5975, 2024, 4911, 2006, 1996, 3509, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09776252 0.14000922 0.17981765 ... 0.38904458 0.17996795 0.3174233 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 3509, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.00539521 0.19177859 0.22123201 ... 0.7693454  0.25675693 0.17193547]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 5660, 2327, 2515, 2023, 3829, 2031, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.00539521 0.19177859 0.22123201 ... 0.7693454  0.25675693 0.17193547]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 4620, 2024, 3491, 2182, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.00539521 0.19177859 0.22123201 ... 0.7693454  0.25675693 0.17193547]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996, 18302,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.00539521 0.19177859 0.22123201 ... 0.7693454  0.25675693 0.17193547]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  2921,  1999,  1996, 10439, 15204,  3401,\n",
            "         2104,  1996,  4675,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.00539521 0.19177859 0.22123201 ... 0.7693454  0.25675693 0.17193547]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2282, 2003, 2023, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.13852791 0.3245231  0.5436648  ... 0.22218053 0.11296794 0.92332083]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 2158, 2183, 2091, 1037, 9663, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.13852791 0.3245231  0.5436648  ... 0.22218053 0.11296794 0.92332083]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 2158, 2725, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.13852791 0.3245231  0.5436648  ... 0.22218053 0.11296794 0.92332083]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2006, 1996, 2158, 1005, 1055, 2398, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18290474 0.4756295  0.41193968 ... 0.19213007 0.39457107 0.16097748]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  2006,  1996, 20619,  2006,  1996, 11851,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18290474 0.4756295  0.41193968 ... 0.19213007 0.39457107 0.16097748]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2071,  1996, 19645,  2022,  2980,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18290474 0.4756295  0.41193968 ... 0.19213007 0.39457107 0.16097748]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054, 19645,  2003,  1999,  1996,  2452,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5329765  0.53454685 0.25353792 ... 0.21636032 0.08524045 0.01436857]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2515, 2023, 7954, 5383, 6240, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5329765  0.53454685 0.25353792 ... 0.21636032 0.08524045 0.01436857]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116,  4127,  1997, 11546,  2024,  2045,  1999,\n",
            "         2023,  3861,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5329765  0.53454685 0.25353792 ... 0.21636032 0.08524045 0.01436857]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2024,  1996, 25659,  2015,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11042964 0.09453127 0.13418885 ... 0.18363616 0.42980555 0.65809005]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2024, 2122, 9738, 3909, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11042964 0.09453127 0.13418885 ... 0.18363616 0.42980555 0.65809005]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2079,  2122, 24042,  2031,  9924,  2030, 23405,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11042964 0.09453127 0.13418885 ... 0.18363616 0.42980555 0.65809005]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2024, 2122, 9738, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.84383094 0.20568585 1.3520631  ... 0.37669435 1.1335034  0.34152168]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003, 10334,  4147,  1037,  3614,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.84383094 0.20568585 1.3520631  ... 0.37669435 1.1335034  0.34152168]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2828, 1997, 2282, 2003, 2023, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.84383094 0.20568585 1.3520631  ... 0.37669435 1.1335034  0.34152168]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 2010, 5495, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1805846  0.27892455 0.18846786 ... 0.03685596 0.12985605 0.35473967]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2630, 4874, 1999, 1996, 3746, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1805846  0.27892455 0.18846786 ... 0.03685596 0.12985605 0.35473967]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2339,  2052,  1996, 12699,  3698,  2022,  1999,  1996,\n",
            "         5723,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1805846  0.27892455 0.18846786 ... 0.03685596 0.12985605 0.35473967]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  1996,  3609,  1997,  1996, 10257,  2006,\n",
            "         1996, 14366,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18356994 0.0419952  0.21939515 ... 0.14668515 0.27223822 0.55885124]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003, 13423,  2041,  1997,  1996,  8494,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18356994 0.0419952  0.21939515 ... 0.14668515 0.27223822 0.55885124]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2339, 2052, 2619, 2681, 2037, 4049, 2006, 1996, 3509, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18356994 0.0419952  0.21939515 ... 0.14668515 0.27223822 0.55885124]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 5568, 1999, 2023, 3861, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18356994 0.0419952  0.21939515 ... 0.14668515 0.27223822 0.55885124]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1999, 1996, 4281, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18356994 0.0419952  0.21939515 ... 0.14668515 0.27223822 0.55885124]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 3154, 2154, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18356994 0.0419952  0.21939515 ... 0.14668515 0.27223822 0.55885124]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[ 101, 2073, 2024, 2027, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18356994 0.0419952  0.21939515 ... 0.14668515 0.27223822 0.55885124]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2064, 1996, 6242, 2175, 7650, 1996, 2958, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18356994 0.0419952  0.21939515 ... 0.14668515 0.27223822 0.55885124]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2515, 1996, 2300, 3711, 2000, 2022, 2659, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.02972686 0.20760539 0.15434912 ... 0.650586   0.1831541  0.39141506]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2003, 2016, 2006, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.02972686 0.20760539 0.15434912 ... 0.650586   0.1831541  0.39141506]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2214, 2003, 2023, 2711, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.02972686 0.20760539 0.15434912 ... 0.650586   0.1831541  0.39141506]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023,  2450, 19448, 12701,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.30581677 0.28506303 0.23681514 ... 0.42162716 0.55929255 0.54492426]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  1999,  1996, 15723,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.30581677 0.28506303 0.23681514 ... 0.42162716 0.55929255 0.54492426]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2008,  1037, 25659,  1999,  2008, 11642,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.30581677 0.28506303 0.23681514 ... 0.42162716 0.55929255 0.54492426]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996, 10457, 24759, 11823,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2645863  0.23065132 0.58037186 ... 0.2921898  0.59118295 0.07457918]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 5723, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2645863  0.23065132 0.58037186 ... 0.2921898  0.59118295 0.07457918]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2073, 2003, 2023, 5723, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2645863  0.23065132 0.58037186 ... 0.2921898  0.59118295 0.07457918]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996, 26783,  1011,  2828, 11848,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.39317688 0.5769734  0.10716923 ... 1.175568   0.2919271  0.59857476]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 7395, 2024, 5710, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.39317688 0.5769734  0.10716923 ... 1.175568   0.2919271  0.59857476]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 3899, 2648, 2030, 2503, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.39317688 0.5769734  0.10716923 ... 1.175568   0.2919271  0.59857476]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 8843, 1997, 3899, 2003, 2008, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.9173057  0.09849977 0.0956926  ... 0.5327665  0.26661873 0.06516334]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 1037, 3137, 1999, 1996, 4281, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.9173057  0.09849977 0.0956926  ... 0.5327665  0.26661873 0.06516334]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2073, 2003, 2023, 2012, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.9173057  0.09849977 0.0956926  ... 0.5327665  0.26661873 0.06516334]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2024,  3564,  2006,  1996, 29145,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.9173057  0.09849977 0.0956926  ... 0.5327665  0.26661873 0.06516334]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996, 29145,  5983,  5568,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.9173057  0.09849977 0.0956926  ... 0.5327665  0.26661873 0.06516334]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 2023, 4111, 2648, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.9173057  0.09849977 0.0956926  ... 0.5327665  0.26661873 0.06516334]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  1996,  2521, 20515,  4111,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6583111  0.41413733 0.6993481  ... 0.12120881 0.8294836  0.2995404 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2023, 1037, 2273, 1005, 1055, 2282, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6583111  0.41413733 0.6993481  ... 0.12120881 0.8294836  0.2995404 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 4084, 2024, 2045, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6583111  0.41413733 0.6993481  ... 0.12120881 0.8294836  0.2995404 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=\n",
            "array([[  101,  2001,  1996,  2197,  2711,  1999,  1996,  5723, 20868,\n",
            "         6072, 26029, 19307,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6583111  0.41413733 0.6993481  ... 0.12120881 0.8294836  0.2995404 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996, 11848, 11876,  2330,  2030,  2701,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6583111  0.41413733 0.6993481  ... 0.12120881 0.8294836  0.2995404 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 7752, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6583111  0.41413733 0.6993481  ... 0.12120881 0.8294836  0.2995404 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2282, 2003, 2023, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6583111  0.41413733 0.6993481  ... 0.12120881 0.8294836  0.2995404 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 2723, 2003, 1999, 2023, 2282, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6583111  0.41413733 0.6993481  ... 0.12120881 0.8294836  0.2995404 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 15), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2515,  1037,  2711,  2079,  2007,  1996,  3165,\n",
            "         3482,  2682,  1996, 11848,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 15), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6583111  0.41413733 0.6993481  ... 0.12120881 0.8294836  0.2995404 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 3309, 5723, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6583111  0.41413733 0.6993481  ... 0.12120881 0.8294836  0.2995404 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2003, 2023, 2282, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6583111  0.41413733 0.6993481  ... 0.12120881 0.8294836  0.2995404 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2045, 1037, 5259, 2182, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6583111  0.41413733 0.6993481  ... 0.12120881 0.8294836  0.2995404 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2502, 2003, 1996, 6457, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6583111  0.41413733 0.6993481  ... 0.12120881 0.8294836  0.2995404 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 20452,  2015,  2024,  2379,  1996,  2341,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6583111  0.41413733 0.6993481  ... 0.12120881 0.8294836  0.2995404 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996, 11848,  2835,  2039,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6583111  0.41413733 0.6993481  ... 0.12120881 0.8294836  0.2995404 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 2341, 2441, 2030, 2701, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6583111  0.41413733 0.6993481  ... 0.12120881 0.8294836  0.2995404 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2052, 2022, 2242, 8250, 1999, 1037, 5239, 2182, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.63006884 0.40230945 0.29150197 ... 0.44485602 0.93633664 0.44355845]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 4874, 2682, 1996, 4937, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.63006884 0.40230945 0.29150197 ... 0.44485602 0.93633664 0.44355845]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  1996,  4937, 10201,  2006,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.63006884 0.40230945 0.29150197 ... 0.44485602 0.93633664 0.44355845]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2106, 1996, 4937, 2079, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.63006884 0.40230945 0.29150197 ... 0.44485602 0.93633664 0.44355845]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 4937, 5777, 2006, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.63006884 0.40230945 0.29150197 ... 0.44485602 0.93633664 0.44355845]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 4937, 2652, 2007, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.63006884 0.40230945 0.29150197 ... 0.44485602 0.93633664 0.44355845]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 4937, 3564, 2006, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.63006884 0.40230945 0.29150197 ... 0.44485602 0.93633664 0.44355845]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2214, 2003, 1996, 4937, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.63006884 0.40230945 0.29150197 ... 0.44485602 0.93633664 0.44355845]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996, 11848,  3259,  4897,  4064,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.02176936 1.020108   0.0086264  ... 0.8024738  0.11912994 0.10027446]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2515,  1996, 10733,  2031, 23827,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.02176936 1.020108   0.0086264  ... 0.8024738  0.11912994 0.10027446]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 5127, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.02176936 1.020108   0.0086264  ... 0.8024738  0.11912994 0.10027446]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 26191,  2024,  2006,  1996, 11345,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.02176936 1.020108   0.0086264  ... 0.8024738  0.11912994 0.10027446]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2045,  2151,  6240,  2006,  1996, 10733,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.02176936 1.020108   0.0086264  ... 0.8024738  0.11912994 0.10027446]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023, 10733,  3201,  2000,  4521,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.02176936 1.020108   0.0086264  ... 0.8024738  0.11912994 0.10027446]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 2795, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.02176936 1.020108   0.0086264  ... 0.8024738  0.11912994 0.10027446]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2038,  1996, 10733,  2042, 15920,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.02176936 1.020108   0.0086264  ... 0.8024738  0.11912994 0.10027446]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  1996,  2665,  8875,  2006,  1996, 10733,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.02176936 1.020108   0.0086264  ... 0.8024738  0.11912994 0.10027446]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2038,  2023, 10733,  2042, 15920,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.02176936 1.020108   0.0086264  ... 0.8024738  0.11912994 0.10027446]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2052,  1037, 23566,  4521,  2023,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.02176936 1.020108   0.0086264  ... 0.8024738  0.11912994 0.10027446]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996,  2878, 10733,  5710,  2306,  1996,  3861,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.02176936 1.020108   0.0086264  ... 0.8024738  0.11912994 0.10027446]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 25609,  2024,  2045,  1997, 10733,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.02176936 1.020108   0.0086264  ... 0.8024738  0.11912994 0.10027446]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023,  1037,  2878, 10733,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.7938205  0.14092904 0.2817538  ... 0.47440565 0.6498971  1.4386064 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2828, 1997, 2311, 2003, 2023, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.7938205  0.14092904 0.2817538  ... 0.47440565 0.6498971  1.4386064 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 2711, 2725, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.7938205  0.14092904 0.2817538  ... 0.47440565 0.6498971  1.4386064 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  1996, 21776,  2725,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3663153  0.46973202 0.09706888 ... 0.44872978 0.04993623 0.37624487]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 4845, 4147, 1037, 3869, 6045, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3663153  0.46973202 0.09706888 ... 0.44872978 0.04993623 0.37624487]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 15), dtype=int32, numpy=\n",
            "array([[ 101, 2515, 1996, 2235, 2879, 1005, 1055, 2606, 2272, 2091, 2000,\n",
            "        2010, 8407, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 15), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.3663153  0.46973202 0.09706888 ... 0.44872978 0.04993623 0.37624487]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 1996, 4845, 5983, 2833, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6116705  0.41913936 0.8971361  ... 0.13407326 0.5337956  0.16298953]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 2111, 2024, 1999, 2317, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6116705  0.41913936 0.8971361  ... 0.13407326 0.5337956  0.16298953]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2653, 2003, 2006, 1996, 4748, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6116705  0.41913936 0.8971361  ... 0.13407326 0.5337956  0.16298953]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 2045, 2062, 2273, 2084, 2308, 1999, 2023, 6302, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.28556326 0.26583886 0.19510306 ... 0.19525273 0.3802957  0.10519168]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 1037, 2958, 1999, 2023, 6302, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.28556326 0.26583886 0.19510306 ... 0.19525273 0.3802957  0.10519168]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2024, 2216, 3628, 2665, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.28556326 0.26583886 0.19510306 ... 0.19525273 0.3802957  0.10519168]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2073, 2003, 2023, 2284, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.28556326 0.26583886 0.19510306 ... 0.19525273 0.3802957  0.10519168]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2024, 1996, 7793, 5153, 1996, 2958, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.28556326 0.26583886 0.19510306 ... 0.19525273 0.3802957  0.10519168]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 1037, 4049, 1999, 1996, 2300, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.28556326 0.26583886 0.19510306 ... 0.19525273 0.3802957  0.10519168]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 2045, 6242, 8274, 1999, 1996, 2300, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.28556326 0.26583886 0.19510306 ... 0.19525273 0.3802957  0.10519168]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3430, 2003, 1996, 8946, 2081, 1997, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.28556326 0.26583886 0.19510306 ... 0.19525273 0.3802957  0.10519168]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116,  2829, 21877, 19341,  3619,  2024,  1999,\n",
            "         1996,  3861,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.28556326 0.26583886 0.19510306 ... 0.19525273 0.3802957  0.10519168]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 8946, 2214, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.28556326 0.26583886 0.19510306 ... 0.19525273 0.3802957  0.10519168]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2161, 2003, 2009, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.28556326 0.26583886 0.19510306 ... 0.19525273 0.3802957  0.10519168]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 2006, 1996, 2958, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.28556326 0.26583886 0.19510306 ... 0.19525273 0.3802957  0.10519168]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 4176, 2024, 2556, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08677861 0.1714449  0.12567924 ... 0.7397139  0.33640754 0.38299826]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2073, 2003, 1996, 4799, 6847, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08677861 0.1714449  0.12567924 ... 0.7397139  0.33640754 0.38299826]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2172, 2936, 2127, 7090, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08677861 0.1714449  0.12567924 ... 0.7397139  0.33640754 0.38299826]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996,  5119, 15056,  5044,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08677861 0.1714449  0.12567924 ... 0.7397139  0.33640754 0.38299826]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 3953, 1997, 1996, 2813, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08677861 0.1714449  0.12567924 ... 0.7397139  0.33640754 0.38299826]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 2396, 2006, 1996, 2813, 1010, 3701, 2417, 2030,\n",
            "        4589, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20814069 0.1338244  0.09279308 ... 0.8444109  0.83180857 0.42191413]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2024,  1996, 14829,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20814069 0.1338244  0.09279308 ... 0.8444109  0.83180857 0.42191413]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2171, 1997, 2010, 6007, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20814069 0.1338244  0.09279308 ... 0.8444109  0.83180857 0.42191413]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 2023, 7954, 7965, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20814069 0.1338244  0.09279308 ... 0.8444109  0.83180857 0.42191413]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2106,  1996,  2158, 27545,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09086668 0.18428579 0.30232373 ... 0.51737916 0.13324854 0.00846026]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116,  2111,  2024,  1999,  1996, 27263,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09086668 0.18428579 0.30232373 ... 0.51737916 0.13324854 0.00846026]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2006, 1996, 2158, 1005, 1055, 2227, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09086668 0.18428579 0.30232373 ... 0.51737916 0.13324854 0.00846026]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 5495, 2003, 1996, 2158, 4147, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09086668 0.18428579 0.30232373 ... 0.51737916 0.13324854 0.00846026]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996,  2158, 12766,  2010,  2219,  2606,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25665155 0.17475325 0.7764819  ... 0.24852145 0.06934568 1.8536102 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 5194, 2024, 2045, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25665155 0.17475325 0.7764819  ... 0.24852145 0.06934568 1.8536102 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2339, 2003, 2045, 1037, 3899, 1999, 1996, 6302, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25665155 0.17475325 0.7764819  ... 0.24852145 0.06934568 1.8536102 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2029, 3257, 2003, 1996, 2067, 3586, 5307, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25665155 0.17475325 0.7764819  ... 0.24852145 0.06934568 1.8536102 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 17188,  2024,  1999,  1996,  6302,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25665155 0.17475325 0.7764819  ... 0.24852145 0.06934568 1.8536102 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2024, 1996, 5194, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25665155 0.17475325 0.7764819  ... 0.24852145 0.06934568 1.8536102 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  1996,  2158, 14906,  2075,  2008,  2003,\n",
            "         5559,  1996,  3586,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25665155 0.17475325 0.7764819  ... 0.24852145 0.06934568 1.8536102 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 2035, 1996, 4176, 1996, 2168, 3609, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25665155 0.17475325 0.7764819  ... 0.24852145 0.06934568 1.8536102 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 3628, 1999, 2023, 3861, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25665155 0.17475325 0.7764819  ... 0.24852145 0.06934568 1.8536102 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[ 101, 2129, 2116, 5194, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.25665155 0.17475325 0.7764819  ... 0.24852145 0.06934568 1.8536102 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 2273, 2024, 2045, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.37689444 0.12393321 0.1146023  ... 0.21428779 0.11073691 0.49493554]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 2311, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.37689444 0.12393321 0.1146023  ... 0.21428779 0.11073691 0.49493554]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2009, 1019, 1024, 4002, 7610, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.37689444 0.12393321 0.1146023  ... 0.21428779 0.11073691 0.49493554]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 2151, 3756, 1999, 1996, 6302, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.37689444 0.12393321 0.1146023  ... 0.21428779 0.11073691 0.49493554]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2828, 1997, 2154, 2003, 2009, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.41517895 0.46247178 0.45412448 ... 0.41146922 0.35356426 0.7028396 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 4111, 2003, 1996, 2611, 9004, 3436, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.41517895 0.46247178 0.45412448 ... 0.41146922 0.35356426 0.7028396 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 3797, 2003, 1996, 2611, 4147, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.41517895 0.46247178 0.45412448 ... 0.41146922 0.35356426 0.7028396 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2210, 2611, 9004, 3436, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.8467401  0.03339011 0.26449183 ... 0.9196954  1.5517026  0.68871754]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2515,  1996,  4937,  1997, 13560,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.8467401  0.03339011 0.26449183 ... 0.9196954  1.5517026  0.68871754]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2003,  1996,  3861, 14231,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.8467401  0.03339011 0.26449183 ... 0.9196954  1.5517026  0.68871754]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 4937, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.8467401  0.03339011 0.26449183 ... 0.9196954  1.5517026  0.68871754]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 4937, 5983, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11288556 0.17423677 0.40177798 ... 0.9150138  0.28505713 0.5411816 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2785,  1997,  3302,  2003,  1996, 17260,  6277,\n",
            "         2121, 10080,  2006,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11288556 0.17423677 0.40177798 ... 0.9150138  0.28505713 0.5411816 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2024,  1996, 24789, 14829,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11288556 0.17423677 0.40177798 ... 0.9150138  0.28505713 0.5411816 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2024, 1996, 2477, 2006, 1996, 2158, 1005, 1055, 5042,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11288556 0.17423677 0.40177798 ... 0.9150138  0.28505713 0.5411816 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996,  2158,  4147,  1037, 10412,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11288556 0.17423677 0.40177798 ... 0.9150138  0.28505713 0.5411816 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2006, 1996, 2604, 2545, 2157, 7223, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11288556 0.17423677 0.40177798 ... 0.9150138  0.28505713 0.5411816 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[  101,  2024,  2122,  2111, 17260, 21172,  1999,  1037,  4318,\n",
            "        11987,  8667,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.11288556 0.17423677 0.40177798 ... 0.9150138  0.28505713 0.5411816 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2024,  2010,  6181, 19586,  2304,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.9044654  0.4073822  0.0857918  ... 0.45799354 0.09735323 0.06445114]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3616, 2024, 2006, 1996, 5080, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.9044654  0.4073822  0.0857918  ... 0.45799354 0.09735323 0.06445114]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 2023, 5080, 2005, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.9044654  0.4073822  0.0857918  ... 0.45799354 0.09735323 0.06445114]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 1037, 2051, 5787, 2000, 2224, 2023, 5581, 8316,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2151753  0.0091897  0.2211054  ... 0.00116181 0.7698003  0.2687146 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2079, 2027, 2298, 2066, 2027, 1005, 2128, 2383, 4569, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2151753  0.0091897  0.2211054  ... 0.00116181 0.7698003  0.2687146 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996, 14175,  6277,  1999,  2392,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2151753  0.0091897  0.2211054  ... 0.00116181 0.7698003  0.2687146 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  2023, 27747,  2725,  2007,  2010,  2519,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2151753  0.0091897  0.2211054  ... 0.00116181 0.7698003  0.2687146 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 3467, 4023, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2151753  0.0091897  0.2211054  ... 0.00116181 0.7698003  0.2687146 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 2450, 1005, 1055, 2606, 4954, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2151753  0.0091897  0.2211054  ... 0.00116181 0.7698003  0.2687146 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2024, 1996, 2111, 1999, 2300, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2151753  0.0091897  0.2211054  ... 0.00116181 0.7698003  0.2687146 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2029, 2711, 2003, 1999, 1996, 2599, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2151753  0.0091897  0.2211054  ... 0.00116181 0.7698003  0.2687146 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2040, 2003, 1999, 1996, 2300, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2151753  0.0091897  0.2211054  ... 0.00116181 0.7698003  0.2687146 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 14175, 15271,  1999,  1996,  2300,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2151753  0.0091897  0.2211054  ... 0.00116181 0.7698003  0.2687146 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2024, 2122, 2048, 2725, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.12718514 0.18740916 0.2267606  ... 0.1123064  0.13905951 0.73460495]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2045, 1037, 2304, 8351, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.12718514 0.18740916 0.2267606  ... 0.1123064  0.13905951 0.73460495]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2038,  2023,  7954,  2042,  7919, 12984,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.12718514 0.18740916 0.2267606  ... 0.1123064  0.13905951 0.73460495]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 8351, 2024, 2045, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.30607325 0.27917933 0.786371   ... 0.02395776 0.5867956  0.7047745 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 2111, 2024, 2045, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.30607325 0.27917933 0.786371   ... 0.02395776 0.5867956  0.7047745 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2028,  1997,  1996,  2308,  4147,  1037, 18865,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.30607325 0.27917933 0.786371   ... 0.02395776 0.5867956  0.7047745 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2024, 2122, 2436, 3667, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.30607325 0.27917933 0.786371   ... 0.02395776 0.5867956  0.7047745 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2515, 3071, 2031, 1037, 2835, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.77112055 0.4029354  0.38651186 ... 0.6022054  0.72036695 0.76280713]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2024, 1996, 4111, 6625, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.77112055 0.4029354  0.38651186 ... 0.6022054  0.72036695 0.76280713]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 5697, 2395, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.77112055 0.4029354  0.38651186 ... 0.6022054  0.72036695 0.76280713]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 17), dtype=int32, numpy=\n",
            "array([[  101,  2097,  1037,  2843,  1997,  1996,  2088,  2298,  2066,\n",
            "         2023,  2044,  1037,  6047,  8957, 15336,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 17), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.77112055 0.4029354  0.38651186 ... 0.6022054  0.72036695 0.76280713]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 4176, 2024, 3491, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.77112055 0.4029354  0.38651186 ... 0.6022054  0.72036695 0.76280713]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2024, 1996, 4303, 4799, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.77112055 0.4029354  0.38651186 ... 0.6022054  0.72036695 0.76280713]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 2181, 2448, 2091, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.77112055 0.4029354  0.38651186 ... 0.6022054  0.72036695 0.76280713]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1999, 1996, 3861, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.77112055 0.4029354  0.38651186 ... 0.6022054  0.72036695 0.76280713]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 3348, 1997, 2122, 4111, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.77112055 0.4029354  0.38651186 ... 0.6022054  0.72036695 0.76280713]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[  101,  2003,  2023, 12217,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.77112055 0.4029354  0.38651186 ... 0.6022054  0.72036695 0.76280713]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2724, 3047, 2182, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.77112055 0.4029354  0.38651186 ... 0.6022054  0.72036695 0.76280713]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  4815,  1996, 17744,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6603962  0.30306232 0.5761595  ... 0.1621093  0.6619632  0.12581018]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2024, 2027, 3788, 2006, 2300, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6603962  0.30306232 0.5761595  ... 0.1621093  0.6619632  0.12581018]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2024, 1996, 2111, 3173, 2682, 2037, 4641, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6603962  0.30306232 0.5761595  ... 0.1621093  0.6619632  0.12581018]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 1996, 2111, 3788, 2006, 2300, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6603962  0.30306232 0.5761595  ... 0.1621093  0.6619632  0.12581018]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996,  9111,  1997,  1996, 12977,  3287,  2030,\n",
            "         2931,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6603962  0.30306232 0.5761595  ... 0.1621093  0.6619632  0.12581018]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 2045, 2151, 3121, 1999, 2023, 3746, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6603962  0.30306232 0.5761595  ... 0.1621093  0.6619632  0.12581018]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  1005,  1055,  1996,  3609,  1997,  1996, 12977,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.07306916 0.5903339  0.07956742 ... 0.61349297 0.25762722 0.25640482]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 2158, 1005, 1055, 5495, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.07306916 0.5903339  0.07956742 ... 0.61349297 0.25762722 0.25640482]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023,  2019,  2840,  1998, 14030,  3573,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.07306916 0.5903339  0.07956742 ... 0.61349297 0.25762722 0.25640482]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2194, 2515, 1996, 2158, 2147, 2005, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.07306916 0.5903339  0.07956742 ... 0.61349297 0.25762722 0.25640482]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 2111, 2024, 2045, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2472465  0.11297777 0.32842243 ... 0.32830775 0.4352423  1.0302329 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 10257, 13304,  2024,  2045,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2472465  0.11297777 0.32842243 ... 0.32830775 0.4352423  1.0302329 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996, 14090,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2472465  0.11297777 0.32842243 ... 0.32830775 0.4352423  1.0302329 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116,  2422, 25548,  2024,  2682,  1996,  5259,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2472465  0.11297777 0.32842243 ... 0.32830775 0.4352423  1.0302329 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2828, 1997, 2282, 2003, 2023, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2472465  0.11297777 0.32842243 ... 0.32830775 0.4352423  1.0302329 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2006, 1996, 4675, 2011, 1996, 7752, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2472465  0.11297777 0.32842243 ... 0.32830775 0.4352423  1.0302329 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 2813, 2317, 2030, 3756, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2472465  0.11297777 0.32842243 ... 0.32830775 0.4352423  1.0302329 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 1996, 4597, 2006, 1999, 2023, 5723, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2472465  0.11297777 0.32842243 ... 0.32830775 0.4352423  1.0302329 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 4874, 2682, 1996, 7752, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2472465  0.11297777 0.32842243 ... 0.32830775 0.4352423  1.0302329 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  1996,  5418,  1997,  1996,  6910, 14090,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2472465  0.11297777 0.32842243 ... 0.32830775 0.4352423  1.0302329 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[  101,  2079,  2017,  2156,  2151, 11868, 18623,  2229,  1999,\n",
            "         2023,  5723,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2472465  0.11297777 0.32842243 ... 0.32830775 0.4352423  1.0302329 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2282, 1997, 1996, 2160, 2003, 2023, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2472465  0.11297777 0.32842243 ... 0.32830775 0.4352423  1.0302329 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[  101,  2054, 27902,  5080,  7719,  2000,  1996,  2187,  1997,\n",
            "         2023,  7752,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.2472465  0.11297777 0.32842243 ... 0.32830775 0.4352423  1.0302329 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2129, 2116, 4597, 2024, 2045, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.60855395 0.44984746 0.09606984 ... 0.27440354 1.3596685  1.048557  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 17188,  2024,  3061,  2039,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.60855395 0.44984746 0.09606984 ... 0.27440354 1.3596685  1.048557  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116,  7516,  2024,  2006,  1996, 17188,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.60855395 0.44984746 0.09606984 ... 0.27440354 1.3596685  1.048557  ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 3609, 1997, 1996, 3430, 2122, 4176, 3061,\n",
            "        2006, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5956249  0.1864441  0.3589869  ... 0.2007119  0.0857321  0.42463315]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996,  4937,  7244,  1996, 12191,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5956249  0.1864441  0.3589869  ... 0.2007119  0.0857321  0.42463315]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996, 12191,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5956249  0.1864441  0.3589869  ... 0.2007119  0.0857321  0.42463315]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2054,  6209,  2003,  1996, 10768,  4179, 12964,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5956249  0.1864441  0.3589869  ... 0.2007119  0.0857321  0.42463315]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2045, 1037, 2529, 1999, 2023, 3746, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5956249  0.1864441  0.3589869  ... 0.2007119  0.0857321  0.42463315]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 4937, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5956249  0.1864441  0.3589869  ... 0.2007119  0.0857321  0.42463315]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2194, 22027,  1996, 11486,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5956249  0.1864441  0.3589869  ... 0.2007119  0.0857321  0.42463315]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2172,  2515,  2023,  4937, 17042,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5956249  0.1864441  0.3589869  ... 0.2007119  0.0857321  0.42463315]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 4937, 2478, 1996, 3274, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.05464705 0.5207535  0.5530335  ... 0.39585888 0.07320512 0.8217925 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 18108,  2024,  2417,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.05464705 0.5207535  0.5530335  ... 0.39585888 0.07320512 0.8217925 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116,  1997,  1996, 18108,  2024,  2069,  2665,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.05464705 0.5207535  0.5530335  ... 0.39585888 0.07320512 0.8217925 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 15), dtype=int32, numpy=\n",
            "array([[  101,  2029,  6910, 18108,  2024,  7541,  2000,  1996,  4950,\n",
            "         2008,  2165,  2023,  3861,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 15), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.05464705 0.5207535  0.5530335  ... 0.39585888 0.07320512 0.8217925 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 5909, 2003, 2023, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.05464705 0.5207535  0.5530335  ... 0.39585888 0.07320512 0.8217925 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2024,  2122, 19794,  3044, 18108,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.27286315 0.7166207  0.30740023 ... 0.7570436  0.07543246 0.58188915]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 5907, 2003, 1996, 2711, 1999, 1996, 6302, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.27286315 0.7166207  0.30740023 ... 0.7570436  0.07543246 0.58188915]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 2158, 1005, 1055, 3797, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.27286315 0.7166207  0.30740023 ... 0.7570436  0.07543246 0.58188915]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  1996,  2158,  2006,  1037, 17260,  6277,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.27286315 0.7166207  0.30740023 ... 0.7570436  0.07543246 0.58188915]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 2158, 1999, 1996, 2250, 2030, 2006, 1996, 2598,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6576009  0.14876184 0.53198165 ... 0.6191494  0.62006104 0.45785183]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 6242, 1054, 1999, 1996, 2300, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6576009  0.14876184 0.53198165 ... 0.6191494  0.62006104 0.45785183]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 6242, 2024, 1999, 1996, 2300, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6576009  0.14876184 0.53198165 ... 0.6191494  0.62006104 0.45785183]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2064, 2022, 2464, 2058, 1996, 2300, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6576009  0.14876184 0.53198165 ... 0.6191494  0.62006104 0.45785183]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2024, 1996, 2111, 3173, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6576009  0.14876184 0.53198165 ... 0.6191494  0.62006104 0.45785183]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 2711, 4755, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6576009  0.14876184 0.53198165 ... 0.6191494  0.62006104 0.45785183]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 3124, 3173, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6576009  0.14876184 0.53198165 ... 0.6191494  0.62006104 0.45785183]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2515, 1996, 4633, 2298, 2066, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6576009  0.14876184 0.53198165 ... 0.6191494  0.62006104 0.45785183]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2051, 1997, 2154, 2003, 2009, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6576009  0.14876184 0.53198165 ... 0.6191494  0.62006104 0.45785183]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2073, 2003, 1996, 4049, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6576009  0.14876184 0.53198165 ... 0.6191494  0.62006104 0.45785183]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2079, 2017, 2156, 1037, 3509, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6576009  0.14876184 0.53198165 ... 0.6191494  0.62006104 0.45785183]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 2119, 1997, 2122, 2111, 6001, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6576009  0.14876184 0.53198165 ... 0.6191494  0.62006104 0.45785183]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 1996, 2158, 3666, 1996, 4049, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6576009  0.14876184 0.53198165 ... 0.6191494  0.62006104 0.45785183]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[  101,  2003,  2009, 24057,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6576009  0.14876184 0.53198165 ... 0.6191494  0.62006104 0.45785183]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2006, 1996, 2598, 2011, 2037, 2519, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6576009  0.14876184 0.53198165 ... 0.6191494  0.62006104 0.45785183]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2081,  1996,  2601,  7516,  1999,  1996, 18921,\n",
            "        16365,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6576009  0.14876184 0.53198165 ... 0.6191494  0.62006104 0.45785183]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 2303, 1997, 2300, 2003, 1999, 1996, 3746,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6576009  0.14876184 0.53198165 ... 0.6191494  0.62006104 0.45785183]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2024,  2048,  1997,  1996,  2111, 28380,  3070,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6576009  0.14876184 0.53198165 ... 0.6191494  0.62006104 0.45785183]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2031,  1996,  2111,  2736, 19967,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.6576009  0.14876184 0.53198165 ... 0.6191494  0.62006104 0.45785183]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2024, 2045, 2151, 3628, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.56987804 0.31729263 0.16343372 ... 0.87081933 0.3788377  0.6980951 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996, 17260,  6277,  2006,  1996,  2598,  2030,\n",
            "         1999,  1996,  2250,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 14), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.56987804 0.31729263 0.16343372 ... 0.87081933 0.3788377  0.6980951 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2079, 2017, 2228, 2002, 2097, 2455, 2023, 7577, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.56987804 0.31729263 0.16343372 ... 0.87081933 0.3788377  0.6980951 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2024, 1996, 8044, 2682, 1996, 2311, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5927535  0.08219559 3.3479185  ... 0.76244974 1.9656937  0.17249966]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2024,  1996,  6546, 12402,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5927535  0.08219559 3.3479185  ... 0.76244974 1.9656937  0.17249966]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 4870, 2024, 2122, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5927535  0.08219559 3.3479185  ... 0.76244974 1.9656937  0.17249966]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 3123, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5927535  0.08219559 3.3479185  ... 0.76244974 1.9656937  0.17249966]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 9185, 2079, 2017, 2156, 2006, 1996, 2723,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5927535  0.08219559 3.3479185  ... 0.76244974 1.9656937  0.17249966]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 1999, 2054, 4292, 2001, 2023, 6302, 2579, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5927535  0.08219559 3.3479185  ... 0.76244974 1.9656937  0.17249966]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 2006, 1996, 2795, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5927535  0.08219559 3.3479185  ... 0.76244974 1.9656937  0.17249966]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 29037,  2024,  1999,  1996,  2415, 18781,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5927535  0.08219559 3.3479185  ... 0.76244974 1.9656937  0.17249966]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 3492, 6512, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5927535  0.08219559 3.3479185  ... 0.76244974 1.9656937  0.17249966]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2024, 2122, 3500, 4870, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5927535  0.08219559 3.3479185  ... 0.76244974 1.9656937  0.17249966]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2024, 1996, 4870, 4142, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5927535  0.08219559 3.3479185  ... 0.76244974 1.9656937  0.17249966]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2079,  2017,  2156,  1037, 16876,  5909,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5927535  0.08219559 3.3479185  ... 0.76244974 1.9656937  0.17249966]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996, 18781,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5927535  0.08219559 3.3479185  ... 0.76244974 1.9656937  0.17249966]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 5061, 4874, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5927535  0.08219559 3.3479185  ... 0.76244974 1.9656937  0.17249966]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[  101,  2129,  2116, 24444,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5927535  0.08219559 3.3479185  ... 0.76244974 1.9656937  0.17249966]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2029, 18781,  2038,  1996,  2087,  2300,  1999,  2009,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5927535  0.08219559 3.3479185  ... 0.76244974 1.9656937  0.17249966]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2024, 1996, 4870, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5927535  0.08219559 3.3479185  ... 0.76244974 1.9656937  0.17249966]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 1996, 4870, 2035, 1996, 2168, 3609, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5927535  0.08219559 3.3479185  ... 0.76244974 1.9656937  0.17249966]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 18), dtype=int32, numpy=\n",
            "array([[  101,  2515,  1996,  2795,  4025,  2000,  2022,  8274,  2030,\n",
            "         2024,  2009,  1005,  1055,  3456,  4415, 10358,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 18), dtype=int32, numpy=\n",
            "array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]],\n",
            "      dtype=int32)>}\n",
            "[0.5927535  0.08219559 3.3479185  ... 0.76244974 1.9656937  0.17249966]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003, 15098,  2105,  1996, 18781,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5927535  0.08219559 3.3479185  ... 0.76244974 1.9656937  0.17249966]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[  101,  2024,  2045,  2151,  2482,  9323,  2015,  1999,  1996,\n",
            "        18781,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5927535  0.08219559 3.3479185  ... 0.76244974 1.9656937  0.17249966]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2003,  1996, 18781,  9603,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5927535  0.08219559 3.3479185  ... 0.76244974 1.9656937  0.17249966]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2024, 1996, 3681, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5927535  0.08219559 3.3479185  ... 0.76244974 1.9656937  0.17249966]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2515, 2023, 2272, 2043, 2017, 2344, 4870, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.5927535  0.08219559 3.3479185  ... 0.76244974 1.9656937  0.17249966]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2024, 2122, 7976, 4870, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08223794 0.36446208 0.19499159 ... 0.2404239  0.34186366 0.4789007 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2064, 2017, 2224, 2023, 8962, 3723, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08223794 0.36446208 0.19499159 ... 0.2404239  0.34186366 0.4789007 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2073, 2003, 1996, 2422, 2746, 2013, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08223794 0.36446208 0.19499159 ... 0.2404239  0.34186366 0.4789007 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  5866,  2055,  1996, 11848,  2956,  1999,\n",
            "         2023,  3861,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08223794 0.36446208 0.19499159 ... 0.2404239  0.34186366 0.4789007 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 4867, 2282, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08223794 0.36446208 0.19499159 ... 0.2404239  0.34186366 0.4789007 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023, 11848,  2284,  1999,  1037,  4825,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.24653608 0.27194405 0.28979567 ... 0.87415344 0.1798583  0.7452649 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996,  9841, 28556,  2121,  2330,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.24653608 0.27194405 0.28979567 ... 0.87415344 0.1798583  0.7452649 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 18), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 2019, 4023, 2008, 2089, 2022, 2864, 1999, 1996,\n",
            "        2282, 2956, 1999, 2023, 3861, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 18), dtype=int32, numpy=\n",
            "array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]],\n",
            "      dtype=int32)>}\n",
            "[0.24653608 0.27194405 0.28979567 ... 0.87415344 0.1798583  0.7452649 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996,  2300, 16018,  1999,  1996,  5572, 11008,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.32485086 0.18392168 0.10150778 ... 0.58308417 0.06148434 0.15038528]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 2308, 2024, 2006, 1996, 6847, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.32485086 0.18392168 0.10150778 ... 0.58308417 0.06148434 0.15038528]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2024, 1996, 2308, 1005, 1055, 6879, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.32485086 0.18392168 0.10150778 ... 0.58308417 0.06148434 0.15038528]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 4860, 3147, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20465103 0.13444304 0.19616826 ... 0.28609014 0.35299087 0.04736779]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 4937, 2559, 2012, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20465103 0.13444304 0.19616826 ... 0.28609014 0.35299087 0.04736779]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2785, 1997, 4937, 2003, 2023, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.20465103 0.13444304 0.19616826 ... 0.28609014 0.35299087 0.04736779]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 4937, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09737813 0.5571132  0.38636273 ... 0.40046495 0.73074603 0.3664826 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2024, 2010, 2159, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09737813 0.5571132  0.38636273 ... 0.40046495 0.73074603 0.3664826 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2002,  4147,  1037, 10412,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09737813 0.5571132  0.38636273 ... 0.40046495 0.73074603 0.3664826 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 17), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3430,  2003,  1996, 11661,  1999,  1996,  2158,\n",
            "         1005,  1055,  2157,  2192,  2081,  1997,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 17), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09737813 0.5571132  0.38636273 ... 0.40046495 0.73074603 0.3664826 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  1996,  2158, 14313,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09737813 0.5571132  0.38636273 ... 0.40046495 0.73074603 0.3664826 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023,  2158,  5821,  1037, 17260,  6277,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09737813 0.5571132  0.38636273 ... 0.40046495 0.73074603 0.3664826 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2024, 1996, 7787, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09737813 0.5571132  0.38636273 ... 0.40046495 0.73074603 0.3664826 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2158, 3061, 2006, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.09737813 0.5571132  0.38636273 ... 0.40046495 0.73074603 0.3664826 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2023, 1037, 2158, 2030, 1037, 2450, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08403815 0.25038368 0.24065793 ... 0.62302136 0.5701875  1.2591636 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2515,  2023,  3861,  2031,  1037,  2300, 10665,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08403815 0.25038368 0.24065793 ... 0.62302136 0.5701875  1.2591636 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 6302, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.08403815 0.25038368 0.24065793 ... 0.62302136 0.5701875  1.2591636 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2024,  1996,  2111, 12701,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26233384 0.41210055 0.22931592 ... 0.55094016 0.6224979  0.5733794 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 12586,  2545,  2024,  2045,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26233384 0.41210055 0.22931592 ... 0.55094016 0.6224979  0.5733794 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2024, 2122, 2081, 2041, 1997, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26233384 0.41210055 0.22931592 ... 0.55094016 0.6224979  0.5733794 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2038,  1996,  2833,  2042, 19803,  2525,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26233384 0.41210055 0.22931592 ... 0.55094016 0.6224979  0.5733794 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  1999,  1996, 12586,  2121,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.26233384 0.41210055 0.22931592 ... 0.55094016 0.6224979  0.5733794 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2828,  1997,  4743, 21429,  2003,  2023,  1029,\n",
            "          102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1478258  0.40277687 0.17997485 ... 0.8697112  0.32998064 0.2594361 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  1996,  2158,  8434,  2006,  1996, 18651,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1478258  0.40277687 0.17997485 ... 0.8697112  0.32998064 0.2594361 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=\n",
            "array([[ 101, 2052, 2017, 3710, 2023, 2833, 2000, 1037, 7501, 3586, 1029,\n",
            "         102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 12), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1478258  0.40277687 0.17997485 ... 0.8697112  0.32998064 0.2594361 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 19), dtype=int32, numpy=\n",
            "array([[ 101, 2339, 2987, 1005, 1056, 1996, 5660, 2224, 2010, 2398, 2000,\n",
            "        2404, 1996, 2833, 2006, 1037, 5127, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 19), dtype=int32, numpy=\n",
            "array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]],\n",
            "      dtype=int32)>}\n",
            "[0.1478258  0.40277687 0.17997485 ... 0.8697112  0.32998064 0.2594361 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2158, 1999, 2630, 2725, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.1478258  0.40277687 0.17997485 ... 0.8697112  0.32998064 0.2594361 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2054,  5907,  2003,  1996,  2711,  6276,  1996, 10733,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.35215795 0.13796167 0.68051517 ... 0.91660774 0.52713436 0.32033843]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2003,  1996, 18302,  2357,  2006,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.35215795 0.13796167 0.68051517 ... 0.91660774 0.52713436 0.32033843]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  3609,  2003,  1996, 10135,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.35215795 0.13796167 0.68051517 ... 0.91660774 0.52713436 0.32033843]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 2111, 1999, 1996, 2282, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.43938047 0.48549408 0.2934081  ... 0.42198977 0.44788745 0.1323961 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2040, 2003, 1996, 2450, 2090, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.43938047 0.48549408 0.2934081  ... 0.42198977 0.44788745 0.1323961 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2024, 2045, 2151, 2336, 1999, 1996, 6302, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.43938047 0.48549408 0.2934081  ... 0.42198977 0.44788745 0.1323961 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 2111, 2024, 1999, 1996, 3861, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18360978 0.00546134 0.24838337 ... 0.63710743 0.32520005 0.12724188]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2023, 3861, 1999, 2304, 1998, 2317, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18360978 0.00546134 0.24838337 ... 0.63710743 0.32520005 0.12724188]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 3909, 1999, 1996, 2250, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18360978 0.00546134 0.24838337 ... 0.63710743 0.32520005 0.12724188]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2003, 2023, 1037, 3509, 6302, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.18360978 0.00546134 0.24838337 ... 0.63710743 0.32520005 0.12724188]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2711, 1999, 2023, 6302, 2725, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.19279298 0.35838765 0.2528416  ... 0.25751293 1.053135   0.12053055]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2023,  1037,  7760, 18471,  3861,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.19279298 0.35838765 0.2528416  ... 0.25751293 1.053135   0.12053055]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2711, 3061, 2006, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.19279298 0.35838765 0.2528416  ... 0.25751293 1.053135   0.12053055]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[  101,  2054, 14709,  2003,  4987,  2000,  1996,  2158,  1005,\n",
            "         1055,  2519,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.4625619  0.21880907 0.5015023  ... 0.61725426 0.9719454  0.5604043 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 3609, 2003, 1996, 3861, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.4625619  0.21880907 0.5015023  ... 0.61725426 0.9719454  0.5604043 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 2158, 2725, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.4625619  0.21880907 0.5015023  ... 0.61725426 0.9719454  0.5604043 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2024, 2027, 2383, 4569, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.4625619  0.21880907 0.5015023  ... 0.61725426 0.9719454  0.5604043 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2003, 1996, 2158, 4147, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.0244052  0.25731075 0.6085806  ... 0.10639718 0.11152779 1.1761005 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 4111, 2003, 2006, 1996, 6456, 3797, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.0244052  0.25731075 0.6085806  ... 0.10639718 0.11152779 1.1761005 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2003,  2009, 11695,  2648,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.0244052  0.25731075 0.6085806  ... 0.10639718 0.11152779 1.1761005 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[ 101, 2079, 2027, 3711, 2000, 2022, 2012, 1037, 7668, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.41751912 0.44236228 0.3020632  ... 0.18084598 1.0146224  0.18293637]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 12977,  2015,  2024,  2045,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.41751912 0.44236228 0.3020632  ... 0.18084598 1.0146224  0.18293637]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116, 12977,  2015,  2024,  1999,  1996,  3861,\n",
            "         1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 11), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.41751912 0.44236228 0.3020632  ... 0.18084598 1.0146224  0.18293637]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 2111, 2024, 1999, 1996, 3861, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.41751912 0.44236228 0.3020632  ... 0.18084598 1.0146224  0.18293637]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 3609, 2003, 1996, 2158, 2007, 1996, 2304, 6879, 4147,\n",
            "        1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 13), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.41751912 0.44236228 0.3020632  ... 0.18084598 1.0146224  0.18293637]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 18), dtype=int32, numpy=\n",
            "array([[  101,  2129,  2116,  6087,  2024,  1999,  1996, 12977,  2108,\n",
            "         2218,  2058,  1996,  2177,  1005,  1055,  4641,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 18), dtype=int32, numpy=\n",
            "array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]],\n",
            "      dtype=int32)>}\n",
            "[0.13270378 0.00194285 0.16183051 ... 0.19865945 0.50257486 0.40543646]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2129,  2116, 27747,  2015,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.13270378 0.00194285 0.16183051 ... 0.19865945 0.50257486 0.40543646]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 2003, 2002, 4147, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.13270378 0.00194285 0.16183051 ... 0.19865945 0.50257486 0.40543646]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[  101,  2040, 12271,  1996,  4400,  1029,   102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.13270378 0.00194285 0.16183051 ... 0.19865945 0.50257486 0.40543646]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2003, 1996, 2300, 2630, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.0516372  0.18433066 1.1592314  ... 0.4397871  0.21892443 0.2786437 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2129, 2116, 7821, 2064, 2017, 2156, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.0516372  0.18433066 1.1592314  ... 0.4397871  0.21892443 0.2786437 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2064, 2017, 2156, 1037, 3392, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[1.0516372  0.18433066 1.1592314  ... 0.4397871  0.21892443 0.2786437 ]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 2051, 2003, 2009, 4760, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.37154585 0.45988348 0.09997413 ... 0.04239101 0.59081006 0.92555195]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 2813, 2081, 1997, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.37154585 0.45988348 0.09997413 ... 0.04239101 0.59081006 0.92555195]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  1996, 11190,  2725,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.37154585 0.45988348 0.09997413 ... 0.04239101 0.59081006 0.92555195]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2003, 2008, 1037, 2103, 1999, 1996, 4281, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.37154585 0.45988348 0.09997413 ... 0.04239101 0.59081006 0.92555195]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2003, 1996, 8638, 2081, 2041, 1997, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.81819093 0.14283937 0.21583545 ... 0.59944355 0.16803826 0.08895358]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2029, 4176, 2024, 2122, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.81819093 0.14283937 0.21583545 ... 0.59944355 0.16803826 0.08895358]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=\n",
            "array([[ 101, 2054, 2060, 4111, 2003, 1999, 1996, 3861, 1029,  102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 10), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.81819093 0.14283937 0.21583545 ... 0.59944355 0.16803826 0.08895358]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[ 101, 2054, 6087, 2024, 1996, 4176, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.81819093 0.14283937 0.21583545 ... 0.59944355 0.16803826 0.08895358]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2054, 4111, 2003, 2023, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.81819093 0.14283937 0.21583545 ... 0.59944355 0.16803826 0.08895358]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[ 101, 2003, 2009, 3467, 1029,  102]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 6), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.81819093 0.14283937 0.21583545 ... 0.59944355 0.16803826 0.08895358]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=\n",
            "array([[  101,  2054,  2003,  1996, 29145,  2725,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 8), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n",
            "[0.92227435 0.23548256 0.69665617 ... 0.00609823 0.3866533  0.28386015]\n",
            "{'input_ids': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=\n",
            "array([[  101,  2003,  2045, 17990,  1999,  1996,  3746,  1029,   102]],\n",
            "      dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(1, 9), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1, 1, 1]], dtype=int32)>}\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# # Pass the preprocessed input data to the pre-trained models to get the predicted answers\n",
        "# bert_outputs = []\n",
        "# for d in encoded_data:\n",
        "#     #efnet_output = tf.keras.layers.Flatten()(d[\"image\"])\n",
        "#     efnet_output = d[\"image\"]\n",
        "#     #efnet_output = tf.reshape(efnet_output, (1, 1, 1, -1))  # Reshape to match the expected shape of the image input\n",
        "#     bert_input = {\"input_ids\": d[\"question\"][\"input_ids\"], \"attention_mask\": d[\"question\"][\"attention_mask\"], \"image\": efnet_output}\n",
        "#     bert_output = bert_model(bert_input)\n",
        "#     bert_outputs.append(tf.argmax(bert_output.logits, axis=-1).numpy()[0])"
      ],
      "metadata": {
        "id": "uJc-v4VYE6ta"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#evaluate model"
      ],
      "metadata": {
        "id": "Q3K764BSmqqB"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from transformers import ViltProcessor, ViltForQuestionAnswering\n",
        "import requests\n",
        "from PIL import Image\n",
        "\n",
        "# prepare image + question\n",
        "url = \"http://images.cocodataset.org/val2017/000000039769.jpg\"\n",
        "image = Image.open(requests.get(url, stream=True).raw)\n",
        "text = \"How many cats are there?\"\n",
        "\n",
        "processor = ViltProcessor.from_pretrained(\"dandelin/vilt-b32-finetuned-vqa\")\n",
        "model = ViltForQuestionAnswering.from_pretrained(\"dandelin/vilt-b32-finetuned-vqa\")\n",
        "\n",
        "# prepare inputs\n",
        "encoding = processor(image, text, return_tensors=\"pt\")\n",
        "\n",
        "# forward pass\n",
        "outputs = model(**encoding)\n",
        "logits = outputs.logits\n",
        "idx = logits.argmax(-1).item()\n",
        "print(\"Predicted answer:\", model.config.id2label[idx])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 258,
          "referenced_widgets": [
            "49e1ccf9c1b4465688d7cb0d579b1407",
            "4033b79a480545e097898a7187dc2ad7",
            "3e44c7714327416a952b807129a0b3bb",
            "eb0d4567906147dcac6ea7c8bfe0a004",
            "99cf8da68dfe45e4ac0c905a51d56702",
            "2be8dce2e1b14c25ab6194ea8058d2bc",
            "3193c4be62ab42c6808cafb01fb752cc",
            "550099681e7b47b584399dcb4f90d9ab",
            "6d23bd90602945218d841e600f03db3e",
            "369bf11acaa34be3a7ac89deb51c7b11",
            "35c10214055045ce8805c82d412a5170",
            "b49ca6508edb4bf28274e8095b1000b9",
            "24853d948ce14530876b4c5a9f185cac",
            "14b6c533183b4851ac0e3a3332f400bd",
            "f42eee1160f3440487b67e443f636388",
            "83af66f65cbf44a9860ead99d0c91c48",
            "95e285812f454193bca7bf2bef1dd9ad",
            "b959a5e4809a4eca87e9a8d71ae1ddf2",
            "e0dd34a5ffed465c87241f556f493606",
            "8f9c5a4231674632a45376ccce0e1140",
            "2c5bd8dcef1b4d8aa9549d119c88b1a0",
            "566d4bb4f15c402cb4e41969e48097dc",
            "c0e41e70e5c94065906a7f58ab1438cc",
            "e6bfef5cb040431997b8078391047ebf",
            "ed682b7678cb49c6adbb675d7497eb7e",
            "42d0f9bd693f4547b0a060786a392525",
            "3911081e82f34606ac986fccbb424839",
            "8b33258fae294d40922819b776247572",
            "f0e6503903014c19b3ac9a2103f106cf",
            "cdffd49ceffe44ee9f430436274d23bb",
            "eb7ee9a281ec41f5bf8145fefb1fae27",
            "6f998e064cb949e4b6f36a058cd12b12",
            "5e27afc0f30c43d7ba017df0d25d6ede",
            "02a77e0692144cd295e8c1fe152062b9",
            "29c113e742f4416992ba0e163979cc83",
            "1a119ae9f75347edb31a9dea4c5a321c",
            "4847264d0fcb4a3a8314be734f751a88",
            "f609de01a1dd44519f87f6eca6085e7b",
            "2f037d15541f421a91e485f8ed220cb0",
            "ecb7ba0d62ea48f9aebc2f39683303e5",
            "a9a7c5db46e941e496d7819d9307edb7",
            "fc0ce5cd0bae47f8a2a5b37d1f106b45",
            "cf03e0e5d23f4167a219263678dfc432",
            "82d2d8cfedd648e89ea02a279bb5910c",
            "6dfe4178e81240bd9075c72f4403c036",
            "239f5f9d3fcf4a89afe5c84073f95a08",
            "1593ec48e067470eaafb7042c09f080e",
            "bf966ae4f44b4ac0a5e372326cae885e",
            "b6dfe9f0f307479bb19a25717b9fb58b",
            "dcb364b9559d4b8aa4866551e78b08ee",
            "63db8f4a848c4de9a3bbf1d40f362ece",
            "3aec61d1873f4611af0a21608f571469",
            "7a146b8611bb46058633d84a3fbda022",
            "9d075cb3c6234c0fa32dd06a43779c62",
            "bc81c55cdce741cb967f7949afc1c0be",
            "5c551e7ab19345c78ef5fce4ad1a5d55",
            "33bf2ec4030444da97e2170a7b495c42",
            "04fa3ac3d16447e9ae327de9335b2916",
            "0252733eae1d4a0ba341dbcc5a2c8385",
            "c7f6c028471343d3be2c1ce7baae98b2",
            "782a30615a9240a9abeacd55e72843ca",
            "f40730cfa9d048138c56348127a97ee0",
            "21cb468281ad427895f348068c3b81a9",
            "1bf03756bde240f486501f817b459221",
            "1da6e6fbba124d9a9eb695f0b5a344c7",
            "fbb31acccd2047f392287729fb93f051",
            "e8ba25378e6b4ae0839a7c896b075358",
            "7c1027bb1bd445539503a58cbb0f1058",
            "31b8b82f3d5c4c89ab8ec6b282271770",
            "3539ccad54ef43ba8618c7cbf7b151f4",
            "1a7fc4ca59ab4de98f0b0505daf84255",
            "de0e88d785e5484798ffff60dd614233",
            "fb07c5f562074c9185fc31b9d8678dab",
            "4cdc159e963c4a528f0a81aa0834aeb2",
            "bd8301c50a9e46c6abd209c16d94795a",
            "fb7aa1bc81a84e92b012fe5195bd8d13",
            "dc849e5575564982b134e736ed0896fa"
          ]
        },
        "id": "kQY6tqmvrGQw",
        "outputId": "6a88dc4c-b305-4f72-c014-253afd51b67e"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading (…)rocessor_config.json:   0%|          | 0.00/251 [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "49e1ccf9c1b4465688d7cb0d579b1407"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading (…)okenizer_config.json:   0%|          | 0.00/320 [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "b49ca6508edb4bf28274e8095b1000b9"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading (…)solve/main/vocab.txt:   0%|          | 0.00/232k [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "c0e41e70e5c94065906a7f58ab1438cc"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading (…)/main/tokenizer.json:   0%|          | 0.00/466k [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "02a77e0692144cd295e8c1fe152062b9"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading (…)cial_tokens_map.json:   0%|          | 0.00/112 [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "6dfe4178e81240bd9075c72f4403c036"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading (…)lve/main/config.json:   0%|          | 0.00/136k [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "5c551e7ab19345c78ef5fce4ad1a5d55"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading pytorch_model.bin:   0%|          | 0.00/470M [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "e8ba25378e6b4ae0839a7c896b075358"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Predicted answer: 2\n"
          ]
        }
      ]
    }
  ]
}