phanviethoang1512/MIMESIS-9B
Full fine-tune vs Qwen/Qwen3.5-9B. Fine-tune: 95% of the weights changed, across 32 of 32 layers.
Tensors changed: lm_head 1/1, Embeddings 1/4, MLP 96/208, Attention 178/332, Norms 67/215
1 derivatives checked
- mradermacher/MIMESIS-9B-GGUF: Same weights, converted