OpenGVLab/InternVL3_5-14B-MPO

Full fine-tune vs OpenGVLab/InternVL3_5-14B-Instruct. Fine-tune: 95% of the weights changed, across 40 of 40 layers.

Tensors changed: Embeddings 4/5, Norms 151/257, MLP 216/222, Attention 256/256, other 0/48, lm_head 1/1

2 derivatives checked

• OpenGVLab/InternVL3_5-14B: Full fine-tune
• OpenGVLab/InternVL3_5-14B-HF: Full fine-tune

On the Hub · Downloads on Model Pulse