Hello, I'm trying to generate a summarization of a large document and followed
https://gpt-index.readthedocs.io/en/latest/guides/usage_pattern.html#customizing-llm-s. For the
num_output
variable, I tried a smaller value (like 500) and a bigger value (4000). I was able to get some decent summarization using the smaller value, however, for the big value, the response is like
I'm sorry, I cannot summarize texts that do not exist in the provided context information.
Does anyone know what is causing this? I assumed using a bigger
num_output
will ouput a longer summary. Thanks