changed named of SECONDARY_LLM to SEVEN_B_LLM_MODEL

config.py

@@ -8,15 +8,18 @@ load_dotenv()
 SQLITE_FILE_NAME = os.getenv('SOURCES_CACHE')
 PERSIST_DIRECTORY = os.getenv('VECTOR_DATABASE_LOCATION')
 EMBEDDING_MODEL = os.getenv("EMBEDDING_MODEL")
-SECONDARY_LLM_MODEL = os.getenv("SECONDARY_LLM_MODEL")
+SEVEN_B_LLM_MODEL = os.getenv("SEVEN_B_LLM_MODEL")
 
 
 db = DataBaseHandler()
 
 db.create_all_tables()
 
-SECONDARY_LLM = HuggingFaceEndpoint(
-        repo_id=SECONDARY_LLM_MODEL, 
+# This model is used for task that a larger model may not need to do
+# as of currently we have been getting MODEL OVERLOADED errors
+# with huggingface
+SEVEN_B_LLM_MODEL = HuggingFaceEndpoint(
+        repo_id=SEVEN_B_LLM_MODEL, 
         temperature=0.1,         # Controls randomness in response generation (lower value means less random)
         max_new_tokens=1024,     # Maximum number of new tokens to generate in responses
         repetition_penalty=1.2,  # Penalty for repeating the same words (higher value increases penalty)

example.env

@@ -25,3 +25,5 @@ EMBEDDING_MODEL="sentence-transformers/distiluse-base-multilingual-cased-v2"
 #EMBEDDING_MODEL="sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2" 
 LLM_MODEL="mistralai/Mixtral-8x7B-Instruct-v0.1"
 LLM_MODEL_ARGS=
+
+SEVEN_B_LLM_MODEL="mistralai/Mistral-7B-Instruct-v0.3"

rag_app/knowledge_base/utils.py

@@ -1,6 +1,6 @@
 from langchain_core.documents import Document
 from chains import generate_document_summary_prompt
-from config import SECONDARY_LLM
+from config import SEVEN_B_LLM_MODEL
 
 
 def generate_document_summaries(
@@ -27,7 +27,7 @@ def generate_document_summaries(
     
     for doc in new_docs:
         
-        genrate_summary_chain = generate_document_summary_prompt | SECONDARY_LLM
+        genrate_summary_chain = generate_document_summary_prompt | SEVEN_B_LLM_MODEL
         summary = genrate_summary_chain.invoke(
             {"document":str(doc.metadata)}
         )