alvarobartt/Mistral-7B-v0.1-ORPO
LoRA-style vs mistralai/Mistral-7B-v0.1. 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 32 of 32 layers; norms, biases and embeddings are untouched.
Tensors changed: Norms 0/65, MLP 96/96, Attention 128/128
2 derivatives checked
- mradermacher/Mistral-7B-v0.1-ORPO-i1-GGUF: Quantization
- mradermacher/Mistral-7B-v0.1-ORPO-GGUF: Same weights, converted