kainatq/quantum-keek-v2-7b
Full fine-tune vs Dongwei/DeepSeek-R1-Distill-Qwen-7B-GRPO. Fine-tune: 100% of the weights changed, across 28 of 28 layers.
Tensors changed: Norms 28/57, MLP 84/84, Attention 196/196
2 derivatives checked
- mradermacher/quantum-keek-v2-7b-i1-GGUF: Quantization
- mradermacher/quantum-keek-v2-7b-GGUF: Same weights, converted