allenai/SAGE-MM-Molmo2-8B-SFT

Full fine-tune vs allenai/Molmo2-8B. Fine-tune: 93% of the weights changed, across 36 of 36 layers.

Tensors changed: Norms 245/245, MLP 172/172, Attention 272/272, Embeddings 3/5, other 11/11, lm_head 1/1

1 derivatives checked