rombodawg/Rombos-LLM-V2.5-Qwen-32b
Partial fine-tune vs Qwen/Qwen2.5-32B-Instruct. Partial change: 84% of the weights changed (attention, lm_head, mlp, norms), in 64 of 64 layers.
Tensors changed: Embeddings 0/1, Norms 55/129, MLP 164/192, Attention 340/448, lm_head 1/1
5 derivatives checked
- bartowski/Replete-LLM-V2.5-Qwen-32b-GGUF: Same weights, converted
- mradermacher/Rombos-LLM-V2.5-Qwen-32b-i1-GGUF: Quantization
- mradermacher/Rombos-LLM-V2.5-Qwen-32b-GGUF: Quantization
- sakaltcommunity/novablast-preview: Partial fine-tune
- SubtleOne/Rombos-Qwen2.5-Writer-32b: Full fine-tune
3 other repos hold byte-for-byte identical weight files.