OpenGVLab/InternVL3_5-2B-MPO

Partial fine-tune vs OpenGVLab/InternVL3_5-2B-Instruct. Partial change: 87% of the weights changed (attention, embeddings, lm_head, mlp, norms), in 28 of 28 layers.

Tensors changed: lm_head 1/1, Embeddings 4/5, Norms 141/209, MLP 180/186, Attention 208/208, other 0/48

1 derivatives checked

On the Hub · Downloads on Model Pulse