nanowhale-100m-base / config.json

Commit History

Re-upload with fixed weights (removed _orig_mod prefix from torch.compile)
382fe22
verified

cmpatino HF Staff commited on

Upload SmolDeepSeek-V4 100M pretrained model (5000 steps on FineWeb-Edu)
6e9a78e
verified

cmpatino HF Staff commited on