OpenGVLab/InternVL3_5-38B-Instruct

Full fine-tune vs OpenGVLab/InternVL3_5-38B-Pretrained. Fine-tune: 100% of the weights changed, across 64 of 64 layers.

Tensors changed: Embeddings 5/5, Norms 437/437, Attention 391/391, other 90/90, MLP 378/378, lm_head 1/1

1 derivatives checked

• OpenGVLab/InternVL3_5-38B-MPO: Full fine-tune

1 other repos hold byte-for-byte identical weight files.

On the Hub · Downloads on Model Pulse