Tweaked the way the prompt is formatted going into the LLM query, to avoid the fine-tuned model giving nonsense answers

src/architectures.py

@@ -456,7 +456,8 @@ class HFInferenceEndpoint(ArchitectureComponent):
             "Authorization": f"Bearer {self.api_token}",
             "Content-Type": "application/json"
         }
-        query_input = f"[INST] <<SYS>> {self.system_prompt} <<SYS>> {request.request} [/INST] "
+        #return f"<s>[INST] <<SYS>>\n{sys_prompt}\n<</SYS>>\n\n{q}[/INST]{a}"
+        query_input = f"<s>[INST] <<SYS>>\n{self.system_prompt}\n<</SYS>>\n\n{request.request}[/INST] "
         payload = {
             "inputs": query_input,
             "parameters": {