reasonwang/SkillGym-Qwen3.5-122B-A10B

Partial fine-tune vs Qwen/Qwen3.5-122B-A10B. Partial change: 87% of the weights changed (attention, lm_head, mlp, norms), in 48 of 48 layers.

Tensors changed: MLP 322/1221, lm_head 1/1, Embeddings 0/5, Attention 316/448, other 0/1, Norms 147/273

1 derivatives checked

• mradermacher/SkillGym-Qwen3.5-122B-A10B-GGUF: Quantization

On the Hub · Downloads on Model Pulse