Charles Koutcheme
Charles Koutcheme
Yes, it seems like it. Do you have any proper workaround? I forked the repo and tried disabling caching by specifying headers but it does not work. I guess the...
Yes, that worked for me! It's also quite straightforward to bring the required changes: ```python for _i in range(n): gen_kwargs["seed"] = _i res = client.text_generation( prompt, details=True, **gen_kwargs, ) #...
> When the prompt exceeds the max_length, the log probabilities for both chosen and rejected turn to NaN. Consider filtering out cases where the prompt is longer than the max_length...