bofenghuang/docto-decision-qwen3.5-2b-fr-v0.1
LoRA-style vs Qwen/Qwen3.5-2B. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, k_proj, o_proj, q_proj, up_proj, v_proj changed, in 23 of 24 layers; norms, biases and embeddings are untouched.
Tensors changed: Attention 24/268, Norms 0/183, Embeddings 0/5, MLP 72/175, other 0/1
1 derivatives checked
- mradermacher/docto-decision-qwen3.5-2b-fr-v0.1-GGUF: Same weights, converted