OpenGVLab/Mini-InternVL2-2B-DA-DriveLM
Full fine-tune vs OpenGVLab/InternVL2-2B. Fine-tune: 91% of the weights changed, across 24 of 24 layers.
Tensors changed: Attention 144/144, Norms 128/145, MLP 174/174, Embeddings 4/5, other 49/49
Full fine-tune vs OpenGVLab/InternVL2-2B. Fine-tune: 91% of the weights changed, across 24 of 24 layers.
Tensors changed: Attention 144/144, Norms 128/145, MLP 174/174, Embeddings 4/5, other 49/49