sensenova/SenseNova-SI-1.4-InternVL3-8B

Full fine-tune vs OpenGVLab/InternVL3-8B. Fine-tune: 96% of the weights changed, across 28 of 28 layers.

Tensors changed: Embeddings 1/5, Norms 57/153, MLP 90/186, Attention 196/292, other 0/48, lm_head 1/1

1 derivatives checked

• sensenova/SenseNova-SI-1.5-InternVL3-8B: Partial fine-tune

On the Hub · Downloads on Model Pulse