chihoonlee10/T3Q-DPO-Mistral-7B
LoRA-style vs liminerity/M7-7b. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, up_proj changed, in 32 of 32 layers; norms, biases and embeddings are untouched.
Tensors changed: MLP 96/96, Attention 0/128, Norms 0/65, Embeddings 0/1, lm_head 0/1
2 derivatives checked
- mradermacher/T3Q-DPO-Mistral-7B-GGUF: Quantization
- SrCh1nask1/X-Machina-7b-slerp-v0.0: Full fine-tune