MaziyarPanahi/calme-2.1-qwen2-72b
LoRA-style vs Qwen/Qwen2-72B-Instruct. 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 80 of 80 layers; norms, biases and embeddings are untouched.
Tensors changed: MLP 240/240, Attention 320/560, Norms 0/161
2 derivatives checked
- mradermacher/calme-2.1-qwen2-72b-i1-GGUF: Quantization
- mradermacher/calme-2.1-qwen2-72b-GGUF: Quantization