wzx111/Qwen3-1.7B-MATH-GDPO
Partial fine-tune vs Qwen/Qwen3-1.7B. Partial change: 82% of the weights changed (attention, mlp, norms), in 28 of 28 layers.
Tensors changed: Embeddings 0/1, Norms 82/113, MLP 84/84, Attention 112/112
2 derivatives checked
- mradermacher/Qwen3-1.7B-MATH-GDPO-i1-GGUF: Quantization
- mradermacher/Qwen3-1.7B-MATH-GDPO-GGUF: Same weights, converted