Xirui1208/readall-readtwice-14b-rl-step15-20260925

Partial fine-tune vs Xirui1208/readall-readtwice-14b-stage-a-step230-20260923. Partial change: 89% of the weights changed (attention, mlp, norms), in 48 of 48 layers.

Tensors changed: Embeddings 0/1, Norms 86/97, MLP 144/144, Attention 335/336, lm_head 0/1

2 derivatives checked

3 other repos hold byte-for-byte identical weight files.

On the Hub · Downloads on Model Pulse