rawsh/mirrorqwen2.5-0.5b-SFT
LoRA-style vs unsloth/Qwen2.5-0.5B. 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 24 of 24 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/49, MLP 72/72, Attention 96/168
5 derivatives checked
- mradermacher/mirrorqwen2.5-0.5b-SFT-i1-GGUF: Quantization
- mradermacher/mirrorqwen2.5-0.5b-SFT-GGUF: Same weights, converted
- rawsh/mirrorqwen2.5-0.5b-SimPO-0: Partial fine-tune
- rawsh/mirrorqwen2.5-0.5b-ORPO-1: Partial fine-tune
- rawsh/mirrorqwen2.5-0.5b-SimPO: Partial fine-tune