OpenGVLab/InternVL3_5-241B-A28B-Instruct

Full fine-tune vs OpenGVLab/InternVL3_5-241B-A28B-Pretrained. Fine-tune: 100% of the weights changed, across 94 of 94 layers.

Tensors changed: Norms 557/557, MLP 1836/1836, Attention 511/511, lm_head 1/1, Embeddings 5/5, other 90/90

1 derivatives checked