Sao10K/14B-Qwen2.5-Freya-x1
Full fine-tune vs Qwen/Qwen2.5-14B. Fine-tune: 100% of the weights changed, across 48 of 48 layers.
Tensors changed: Embeddings 1/1, Norms 97/97, MLP 144/144, Attention 336/336, lm_head 1/1
7 derivatives checked
- bartowski/14B-Qwen2.5-Freya-x1-GGUF: Same weights, converted
- mradermacher/14B-Qwen2.5-Freya-x1-i1-GGUF: Quantization
- featherless-ai-quants/Sao10K-14B-Qwen2.5-Freya-x1-GGUF: Quantization
- mradermacher/14B-Qwen2.5-Freya-x1-GGUF: Quantization
- DevQuasar-6/Sao10K.14B-Qwen2.5-Freya-x1-GGUF: Quantization
- MaziyarPanahi/14B-Qwen2.5-Freya-x1-GGUF: Same weights, converted
- Sorawiz/Qwen2.5-14B-Instinct-Talk: Full fine-tune