Ammonix/AmmonixRCM-Writer-9B
LoRA-style vs ornith-ai/Ornith-1.5-9B. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, in_proj_qkv, in_proj_z, k_proj, o_proj, out_proj, q_proj, up_proj, v_proj changed, in 32 of 32 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/4, Norms 0/215, Attention 104/332, MLP 96/208, lm_head 0/1
1 derivatives checked
- mradermacher/AmmonixRCM-Writer-9B-GGUF: Same weights, converted