nicoboss/DeepSeek-R1-Distill-Qwen-32B-Uncensored
LoRA-style vs deepseek-ai/DeepSeek-R1-Distill-Qwen-32B. 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 64 of 64 layers; norms, biases and embeddings are untouched.
Tensors changed: Attention 256/448, MLP 192/192, Norms 0/129
3 derivatives checked
- mradermacher/DeepSeek-R1-Distill-Qwen-32B-Uncensored-GGUF: Quantization
- mradermacher/DeepSeek-R1-Distill-Qwen-32B-Uncensored-i1-GGUF: Quantization
- Riyan200324200324/DeepSeek-R1-Distill-Qwen-32B-Uncensored-GGUF: Quantization
1 other repos hold byte-for-byte identical weight files.