OpenGVLab/InternVL2_5-1B-MPO

Full fine-tune vs OpenGVLab/InternVL2_5-1B. Fine-tune: 100% of the weights changed, across 24 of 24 layers.

Tensors changed: lm_head 1/1, Embeddings 5/5, Norms 68/145, MLP 172/174, Attention 258/264, other 0/48

2 derivatives checked

• mradermacher/InternVL2_5-1B-MPO-GGUF: Same weights, converted (fewer layers)
• mradermacher/InternVL2_5-1B-MPO-i1-GGUF: Quantization (fewer layers)

On the Hub · Downloads on Model Pulse