Minor updates on the model support in doc string post demo cleanup.

PiperOrigin-RevId: 644039366

lit_nlp/examples/blank_slate_demo.py

@@ -6,12 +6,6 @@
 - classification model on MultiNLI, with the MultiNLI dataset.
 - TensorFlow Keras model for penguin classification, with the Penguin tabular
   dataset from TFDS.
-- T5 models using HuggingFace Transformers and Keras, with the English CNNDM
-  summarization dataset and the WMT '14 machine-translation dataset.
-- BERT (bert-*) as a masked language model and GPT-2 (gpt2* or distilgpt2) as a
-  left-to-right language model, with the Stanford Sentiment Treebank dataset,
-  the IMDB reviews dataset, Billion Word Benchmark (lm1b) dataset and the option
-  to load sentences from a flat text file.
 
 To run:
   python -m lit_nlp.examples.blank_slate_demo --port=5432

lit_nlp/examples/prompt_debugging/transformers_lms.py

@@ -1,6 +1,6 @@
 """Wrapper for HuggingFace models in LIT.
 
-Includes BERT masked LM, GPT-2, and T5.
+Supported models include Gemma, GPT-2, Llama, Mistral, etc.
 
 This wrapper loads a model into memory and implements the a number of helper
 functions to predict a batch of examples and extract information such as