Update summ.py

summ.py

@@ -9,7 +9,7 @@ from langchain.chains.combine_documents.stuff import StuffDocumentsChain
 
 os.environ['OPENAI_API_KEY'] = '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'
 
-llm = ChatOpenAI(temperature=0, model_name="gpt-4-1106-preview")
+llm = ChatOpenAI(temperature=0, model_name="gpt-4-0125-preview")
 
 def get_summ(path):
 
@@ -27,6 +27,7 @@ def get_summ(path):
     reduce_template = """The following is set of summaries:
     {doc_summaries}
     Take these and distill it into a final, consolidated summary of the main themes. 
+    Determine the genes relevant or irrelevant to the discussed disease, and incorporate any associated p-values if available.
     Helpful Answer:"""
     reduce_prompt = PromptTemplate.from_template(reduce_template)
 
@@ -45,7 +46,7 @@ def get_summ(path):
         # If documents exceed context for `StuffDocumentsChain`
         collapse_documents_chain=combine_documents_chain,
         # The maximum number of tokens to group documents into.
-        token_max=64000,
+        token_max=100000,
     )
 
     # Combining documents by mapping a chain over them, then combining results
@@ -61,7 +62,7 @@ def get_summ(path):
     )
 
     text_splitter = CharacterTextSplitter.from_tiktoken_encoder(
-        chunk_size=64000, chunk_overlap=0
+        chunk_size=100000, chunk_overlap=0
     )
     split_docs = text_splitter.split_documents(docs)