how to handle context_size token error

[lakshminarayanannn]

Even after setting or increasing the llm context window and max_new_token to max extent in mistral 7b instruct

I get this error

can anyone help me to resolve this.

---

ValueError Traceback (most recent call last)
Cell In[8], line 1
----> 1 res = kg_index.as_query_engine().query(query)
2 print(res)

File ~/LLM_DEV/llm/lib64/python3.9/site-packages/llama_index/core/base_query_engine.py:40, in BaseQueryEngine.query(self, str_or_query_bundle)
38 if isinstance(str_or_query_bundle, str):
39 str_or_query_bundle = QueryBundle(str_or_query_bundle)
---> 40 return self._query(str_or_query_bundle)

File ~/LLM_DEV/llm/lib64/python3.9/site-packages/llama_index/query_engine/retriever_query_engine.py:172, in RetrieverQueryEngine._query(self, query_bundle)
168 with self.callback_manager.event(
169 CBEventType.QUERY, payload={EventPayload.QUERY_STR: query_bundle.query_str}
170 ) as query_event:
171 nodes = self.retrieve(query_bundle)
--> 172 response = self._response_synthesizer.synthesize(
173 query=query_bundle,
174 nodes=nodes,
175 )
177 query_event.on_end(payload={EventPayload.RESPONSE: response})
179 return response

File ~/LLM_DEV/llm/lib64/python3.9/site-packages/llama_index/response_synthesizers/base.py:168, in BaseSynthesizer.synthesize(self, query, nodes, additional_source_nodes, **response_kwargs)
163 query = QueryBundle(query_str=query)
...
152 " not non-negative."
153 )
154 return context_size_tokens

ValueError: Calculated available context size -22 was not non-negative.

[lakshminarayanannn]

@logan-markewich
@jerryjliu

could you please help with this problem .

[logan-markewich]

@lakshminarayanannn would help if you shared some setup code you did to reproduce the issue