OpenGVLab/InternVL3_5-38B-Flash
Full fine-tune vs OpenGVLab/InternVL3_5-38B. Fine-tune: 100% of the weights changed, across 64 of 64 layers.
Tensors changed: Attention 391/391, Norms 351/437, MLP 377/378, Embeddings 4/5, other 90/90, lm_head 1/1
Full fine-tune vs OpenGVLab/InternVL3_5-38B. Fine-tune: 100% of the weights changed, across 64 of 64 layers.
Tensors changed: Attention 391/391, Norms 351/437, MLP 377/378, Embeddings 4/5, other 90/90, lm_head 1/1