netease-youdao/Confucius-o1-14B
Full fine-tune vs Qwen/Qwen2.5-14B-Instruct. Fine-tune: 95% of the weights changed, across 48 of 48 layers.
Tensors changed: Embeddings 0/1, Norms 95/97, MLP 144/144, Attention 336/336, lm_head 1/1
4 derivatives checked
- bartowski/Confucius-o1-14B-GGUF: Same weights, converted
- mradermacher/Confucius-o1-14B-i1-GGUF: Quantization
- netease-youdao/Confucius-o1-14B-GGUF: Same weights, converted
- mradermacher/Confucius-o1-14B-GGUF: Quantization