internlm/OREAL-DeepSeek-R1-Distill-Qwen-7B
Full fine-tune vs deepseek-ai/DeepSeek-R1-Distill-Qwen-7B. Fine-tune: 90% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 1/1, Attention 193/196, MLP 81/84, Norms 37/57, lm_head 0/1
6 derivatives checked
- bartowski/internlm_OREAL-DeepSeek-R1-Distill-Qwen-7B-GGUF: Quantization
- lmstudio-community/OREAL-DeepSeek-R1-Distill-Qwen-7B-GGUF: Quantization
- mradermacher/OREAL-DeepSeek-R1-Distill-Qwen-7B-i1-GGUF: Quantization
- mradermacher/OREAL-DeepSeek-R1-Distill-Qwen-7B-GGUF: Same weights, converted
- tensorblock/OREAL-DeepSeek-R1-Distill-Qwen-7B-GGUF: Quantization
- DevQuasar-7/internlm.OREAL-DeepSeek-R1-Distill-Qwen-7B-GGUF: Same weights, converted