Witold Wydmański

Results 3 issues of Witold Wydmański

GPT-4 and GPT-4o often prepend the json with ` ```json ` text. We now strip it during json loading

Gemma-2 no longer requires `flashinfer` - in fact, newest version of vllm has a bug in its usage, which makes the LLM return wrong tokens. This pull requests makes it...

Hi, the link to preprocessed datasets is not working.