DeryFerd/Qwen2.5-Math-Coder-Distill-Phi-2-4.4K-MixMathCode
LoRA-style vs microsoft/phi-2. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of dense, k_proj, q_proj, v_proj changed, in 32 of 32 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/66, MLP 0/128, Attention 128/256, lm_head 0/2
1 derivatives checked
- mradermacher/Qwen2.5-Math-Coder-Distill-Phi-2-4.4K-MixMathCode-GGUF: Same weights, converted