Ma7ee7/Qwen3.8_4B_Distilled
LoRA-style vs Qwen/Qwen3-4B-Thinking-2507. 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 36 of 36 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/145, MLP 108/108, Attention 144/144
3 derivatives checked
- Ma7ee7/Qwen3.8_4B_Distilled_GGUF: Quantization
- Atomic-Germ/Qwen3.8-Distilled-4B-NPU2: Not comparable yet
- xJoePec/checkpoint-8000: LoRA-style
1 other repos hold byte-for-byte identical weight files.