Zigeng/DMax-Math-16B

Full fine-tune vs inclusionAI/LLaDA2.0-mini. Fine-tune: 92% of the weights changed, across 20 of 20 layers.

Tensors changed: Attention 40/40, Norms 73/81, MLP 2870/2877, Embeddings 0/1, lm_head 1/1

1 derivatives checked

• yresearch/DMax-Math-MMD: Partial fine-tune

On the Hub · Downloads on Model Pulse