OpenGVLab/InternVL2_5-4B-MPO

Full fine-tune vs OpenGVLab/InternVL2_5-4B. Fine-tune: 100% of the weights changed, across 36 of 36 layers.

Tensors changed: Embeddings 5/5, Norms 117/169, MLP 208/210, Attention 348/348, other 0/48, lm_head 1/1

2 derivatives checked

• mradermacher/InternVL2_5-4B-MPO-GGUF: Same weights, converted (fewer layers)
• mradermacher/InternVL2_5-4B-MPO-i1-GGUF: Quantization (fewer layers)

On the Hub · Downloads on Model Pulse