Who has a summary of the game plan for handling aip_requestor.py and memory token overruns? I have not had any long term (more than 10 minutes) run go by without exhausting token limit and crashing instead of trying to automatically handle the exception or error and retry by auto condensin...
for i, token in enumerate(tokens[:-1]): current_token = token next_token = tokens[i + 1] # Check if there's a possible break in the context if some_condition_to_detect_context_break(current_token, next_token): improved_token = generate_more_coherent_token(current_token, next_token)...
hello i am getting this error any idea to fix this : 2023-05-02 16:58:57.051 label got an empty value. This is discouraged for accessibility reasons and may be disallowed in the future by raising an exception. Please provide a non-empty ...
i've already set the token limit to 4000 since i am on GPT3.5, but it's not working, so idk. ### LLM MODEL SETTINGS ## FAST_TOKEN_LIMIT - Fast token limit for OpenAI (Default: 4000) ## SMART_TOKEN_LIMIT - Smart token limit for OpenAI (Default: 8000) ## When using --gpt3on...
(model="gpt-3.5-turbo") DEBUG:guidance.models._remote:finish Remote.__init__ DEBUG:guidance.models._model:start Model._run_stateless DEBUG:guidance.models._remote:start Remote._get_logits(token_ids=[27, 97236, 76, 5011, 91, 29, 882, 198, 38195, 3373, 33894, 27, 97236, 76, 6345,...
I added langchain.debug = True, I saw the requested information, but did not see the returned information, probably because the token limit was exceeded. But there are two strange phenomena, the first one is why there are so many input information? The second is why the prefix of the out...