raghu1155/DeepSeek-R1-Codegeneration-COT
Adapter vs deepseek-ai/DeepSeek-R1. raghu1155/DeepSeek-R1-Codegeneration-COT is a LoRA adapter, not full weights: rank 16, on down_proj, gate_proj, k_proj, o_proj, q_proj, up_proj, v_proj in 32 layers (224 adapted matrices). It modifies deepseek-ai/DeepSeek-R1 at load time.