CewEhao/OPD-Aha-9B
Partial fine-tune vs Qwen/Qwen3.5-9B. Partial change: 89% of the weights changed (attention, embeddings, lm_head, mlp, norms), in 32 of 32 layers.
Tensors changed: lm_head 1/1, Embeddings 3/4, MLP 191/208, Attention 286/332, Norms 152/215
2 derivatives checked
- mradermacher/OPD-Aha-9B-i1-GGUF: Quantization
- mradermacher/OPD-Aha-9B-GGUF: Same weights, converted