OpenGVLab/VisualPRM-8B
Full fine-tune vs OpenGVLab/InternVL2_5-8B. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Attention 160/160, Norms 84/161, MLP 197/198, Embeddings 5/5, other 1/49
Full fine-tune vs OpenGVLab/InternVL2_5-8B. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Attention 160/160, Norms 84/161, MLP 197/198, Embeddings 5/5, other 1/49