readme updates

LLMNeedleHaystackTester.py

@@ -446,7 +446,7 @@ def get_results(self):
 
     def print_start_test_summary(self):
         print ("\n")
-        print ("Starting Needle In A Haystack Testing")
+        print ("Starting Needle In A Haystack Testing...")
         print (f"- Model: {self.model_name}")
         print (f"- Context Lengths: {len(self.context_lengths)}, Min: {min(self.context_lengths)}, Max: {max(self.context_lengths)}")
         print (f"- Document Depths: {len(self.document_depth_percents)}, Min: {min(self.document_depth_percents)}%, Max: {max(self.document_depth_percents)}%")
@@ -460,6 +460,6 @@ def start_test(self):
 
 if __name__ == "__main__":
     # Tons of defaults set, check out the LLMNeedleHaystackTester's init for more info
-    ht = LLMNeedleHaystackTester(document_depth_percent_interval_type='sigmoid', model_provider='Anthropic', model_name='claude-2.1')
+    ht = LLMNeedleHaystackTester()
 
     ht.start_test()

README.md

@@ -6,7 +6,9 @@ A simple 'needle in a haystack' analysis to test in-context retrieval ability of
 
 Get the behind the scenes on the [overview video](https://youtu.be/KwRRuiCCdmc).
 
-**The Test**
+![GPT-4-128 Context Testing](img/NeedleHaystackCodeSnippet.png)
+
+## The Test
 1. Place a random fact or statement (the 'needle') in the middle of a long context window (the 'haystack')
 2. Ask the model to retrieve this statement
 3. Iterate over various document depths (where the needle is placed) and context lengths to measure performance

requirements.txt

@@ -0,0 +1,45 @@
+aiohttp==3.9.1
+aiosignal==1.3.1
+annotated-types==0.6.0
+anthropic==0.7.5
+anyio==3.7.1
+attrs==23.1.0
+certifi==2023.11.17
+charset-normalizer==3.3.2
+dataclasses-json==0.6.3
+distro==1.8.0
+filelock==3.13.1
+frozenlist==1.4.0
+fsspec==2023.10.0
+h11==0.14.0
+httpcore==1.0.2
+httpx==0.25.2
+huggingface-hub==0.19.4
+idna==3.6
+jsonpatch==1.33
+jsonpointer==2.4
+langchain==0.0.341
+langchain-core==0.0.6
+langsmith==0.0.67
+marshmallow==3.20.1
+multidict==6.0.4
+mypy-extensions==1.0.0
+numpy==1.26.2
+openai==1.3.5
+packaging==23.2
+pydantic==2.5.2
+pydantic_core==2.14.5
+python-dotenv==1.0.0
+PyYAML==6.0.1
+regex==2023.10.3
+requests==2.31.0
+sniffio==1.3.0
+SQLAlchemy==2.0.23
+tenacity==8.2.3
+tiktoken==0.5.1
+tokenizers==0.15.0
+tqdm==4.66.1
+typing-inspect==0.9.0
+typing_extensions==4.8.0
+urllib3==2.1.0
+yarl==1.9.3