OpenGVLab/InternVL3-38B-hf
Full fine-tune vs OpenGVLab/InternVL3-38B-Instruct. Fine-tune: 100% of the weights changed, across 64 of 64 layers.
Tensors changed: Embeddings 1/1, Norms 125/129, MLP 192/192, Attention 446/448, lm_head 1/1
Full fine-tune vs OpenGVLab/InternVL3-38B-Instruct. Fine-tune: 100% of the weights changed, across 64 of 64 layers.
Tensors changed: Embeddings 1/1, Norms 125/129, MLP 192/192, Attention 446/448, lm_head 1/1