nbeerbower/bophades-mistral-math-DPO-7B
LoRA-style vs nbeerbower/bophades-v2-mistral-7B. 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: lm_head 0/1, Norms 0/65, MLP 96/96, Attention 128/128, Embeddings 0/1
3 derivatives checked
- featherless-ai-quants/nbeerbower-bophades-mistral-math-DPO-7B-GGUF: Quantization
- nbeerbower/HeroBophades-2x7B: MoE built from dense models
- nbeerbower/slerp-bophades-truthy-math-mistral-7B: LoRA-style