stanford-crfm / mistral

Mistral: A strong, northwesterly wind: Framework for transparent and accessible large-scale language model training, built with Hugging Face 🤗 Transformers.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Generate Model Cards for models

dlwh opened this issue · comments

HF's trainer supports auto-generating a model card given some metadata. We should use this, and add flags/hooks to provide whatever else is missing. Probably should also support push_to_hub while we're at it.

Additional context

See also #86 . We should think about how and if we want to share multiple checkpoints per run automatically.

Eh, whatever