Zhihu-ai/Zhi-Create-DSR1-14B
Full fine-tune vs deepseek-ai/DeepSeek-R1-Distill-Qwen-14B. Fine-tune: 95% of the weights changed, across 48 of 48 layers.
Tensors changed: Embeddings 0/1, Attention 336/336, MLP 144/144, Norms 97/97, lm_head 1/1
5 derivatives checked
- mradermacher/Zhi-Create-DSR1-14B-i1-GGUF: Quantization
- mradermacher/Zhi-writing-dsr1-14b-i1-GGUF: Quantization
- mradermacher/Zhi-Create-DSR1-14B-GGUF: Quantization
- mradermacher/Zhi-writing-dsr1-14b-GGUF: Quantization
- DevQuasar/Zhihu-ai.Zhi-writing-dsr1-14b-GGUF: Same weights, converted