theprint/GeneralChat-Llama3.2-3B-DPO
LoRA-style vs theprint/GeneralChat-Llama3.2-3B. 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 28 of 28 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/57, MLP 84/84, Attention 112/112
3 derivatives checked
- theprint/GeneralChat-Llama3.2-3B-DPO-GGUF: Same weights, converted
- mradermacher/GeneralChat-Llama3.2-3B-DPO-i1-GGUF: Quantization
- mradermacher/GeneralChat-Llama3.2-3B-DPO-GGUF: Same weights, converted