suayptalha/Maestro-R1-Llama-8B
LoRA-style vs deepseek-ai/DeepSeek-R1-Distill-Llama-8B. 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: Embeddings 0/1, Attention 128/128, MLP 96/96, Norms 0/65, lm_head 0/1
2 derivatives checked
- mradermacher/Maestro-R1-Llama-8B-i1-GGUF: Quantization
- mradermacher/Maestro-R1-Llama-8B-GGUF: Same weights, converted