rombodawg/Rombos-LLM-V2.6-Qwen-14b
Full fine-tune vs Qwen/Qwen2.5-14B-Instruct. Fine-tune: 100% of the weights changed, across 48 of 48 layers.
Tensors changed: Embeddings 1/1, Norms 86/97, MLP 144/144, Attention 335/336, lm_head 1/1
6 derivatives checked
- bartowski/Rombos-LLM-V2.6-Qwen-14b-GGUF: Same weights, converted
- mradermacher/Rombos-LLM-V2.6-Qwen-14b-i1-GGUF: Quantization
- featherless-ai-quants/rombodawg-Rombos-LLM-V2.6-Qwen-14b-GGUF: Quantization
- mradermacher/Rombos-LLM-V2.6-Qwen-14b-GGUF: Quantization
- Xiaojian9992024/Qwen2.5-THREADRIPPER-Medium-Censored: Full fine-tune
- synergetic/FrankenQwen2.5-14B: Partial fine-tune