littlemex/prismyra-decision-qwen3.6-35b-a3b-fp8-36l

Partial fine-tune vs Qwen/Qwen3.6-35B-A3B-FP8. Partial change: 25% of the weights changed (attention, mlp), in 36 of 36 layers.

Tensors changed: Norms 0/234, Attention 234/485, MLP 216/2024, other 0/1, Embeddings 0/5, lm_head 0/1

On the Hub · Downloads on Model Pulse