bunnycore/Qwen-2.5-7B-R1-Stock
Full fine-tune vs Qwen/Qwen2.5-7B-Instruct. Fine-tune: 100% of the weights changed, across 28 of 28 layers.
Tensors changed: Norms 2/57, MLP 84/84, Attention 114/196
2 derivatives checked
- mradermacher/Qwen-2.5-7B-R1-Stock-i1-GGUF: Quantization
- mradermacher/Qwen-2.5-7B-R1-Stock-GGUF: Same weights, converted