stevensama73/Qwen2.5-3B-grpo-indonesian
LoRA-style vs stevensama73/Qwen2.5-3B-sft-think-indonesian. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of k_proj, o_proj, q_proj, v_proj changed, in 36 of 36 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/73, MLP 0/108, Attention 144/252
1 derivatives checked
- mradermacher/Qwen2.5-3B-grpo-indonesian-GGUF: LoRA-style