allenai/SAGE-MM-Molmo2-8B-SFT_RL
Partial fine-tune vs allenai/SAGE-MM-Molmo2-8B-SFT. Partial change: 87% of the weights changed (attention, lm_head, mlp, norms), in 36 of 36 layers.
Tensors changed: Norms 94/245, MLP 72/172, Attention 72/272, Embeddings 0/5, lm_head 1/1, other 0/11