OpenGVLab/InternVL3_5-8B-MPO

Full fine-tune vs OpenGVLab/InternVL3_5-8B-Instruct. Fine-tune: 93% of the weights changed, across 36 of 36 layers.

Tensors changed: Embeddings 4/5, Norms 147/241, MLP 203/210, Attention 240/240, other 0/48, lm_head 1/1

2 derivatives checked

• OpenGVLab/InternVL3_5-8B: Partial fine-tune
• OpenGVLab/InternVL3_5-8B-HF: Full fine-tune

On the Hub · Downloads on Model Pulse