v000000/Qwen2.5-Lumen-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 17/97, MLP 144/144, Attention 230/336, lm_head 1/1
4 derivatives checked
- mradermacher/Qwen2.5-Lumen-14B-i1-GGUF: Quantization
- mradermacher/Qwen2.5-Lumen-14B-GGUF: Quantization
- featherless-ai-quants/v000000-Qwen2.5-Lumen-14B-GGUF: Quantization
- tensorblock/Qwen2.5-Lumen-14B-GGUF: Quantization
1 other repos hold byte-for-byte identical weight files.