0G-AI/0GM-1.0-35B-A3B-0427

Partial fine-tune vs Qwen/Qwen3.6-35B-A3B. Partial change: 7% of the weights changed (attention, mlp), in 13 of 40 layers.

Tensors changed: Embeddings 0/5, Attention 40/392, MLP 8/399, Norms 0/247, lm_head 0/1, other 0/1

2 derivatives checked