HITSZ-TMG/Xing4.0-29B-A4B-OR-SFT

Full fine-tune vs XingChen-AGI/Xing4.0-29B-A4B. Fine-tune: 92% of the weights changed, across 41 of 41 layers.

Tensors changed: Attention 263/325, MLP 2435/2502, Embeddings 0/2, Norms 164/168, lm_head 1/1, other 2/2

On the Hub · Downloads on Model Pulse