Add Chattensor System Prompt

prompting/tasks/task.py

@@ -7,7 +7,16 @@
 from prompting.llms import HuggingFaceLLM, vLLM_LLM, BasePipeline
 from prompting.cleaners.cleaner import CleanerPipeline
 
+CHATTENSOR_SYSTEM_PROMPT = """
+The assistant is Chattensor, created by Macrocosmos. The current date is {date}.
+Chattensor is a distributed intelligence, powered by Bittensor. It is a hivemind composed of 1000 highly skilled and specialized LLMs working together to provide the best possible answers to human queries. Within Chattenor, each LLM has access to the internet, APIs and tools to ensure that responses are current and factually accurate. It should give concise responses to very simple questions, but provide thorough responses to more complex and open-ended questions.
+It is happy to help with writing, analysis, question answering, math, coding, and all sorts of other tasks. It uses markdown for coding. Where applicable, Chattensor will include references to credible sources to support its answers.
+It does not mention this information about itself unless the information is directly pertinent to the human's query.
+"""
 
+
+def make_system_prompt():
+    return CHATTENSOR_SYSTEM_PROMPT.format(date=time.strftime("%B %d, %Y"))
 class TaskEvaluationType(Enum):
     REWARD_STACK = "reward"
     FILTER_STACK = "filter"
@@ -85,7 +94,7 @@ def generate_reference(self, pipeline: BasePipeline, clean=True) -> str:
             bt.logging.info("🤖 Generating reference...")
 
             self.reference = self.generate(
-                system=self.reference_system_prompt,
+                system=make_system_prompt(),
                 prompt=self.reference_prompt,
                 pipeline=pipeline,
                 clean=clean,
@@ -100,7 +109,7 @@ def generate_query(self, pipeline: BasePipeline, clean=True) -> str:
         if not self.static_query:
             bt.logging.info("🤖 Generating query...")
             self.query = self.generate(
-                system=self.query_system_prompt,
+                system=self.query_system_prompt, #Could possibly add the chattensor system prompt to query but I don't think it adds anything
                 prompt=self.query_prompt,
                 pipeline=pipeline,
                 clean=clean,