bunnycore/Qwen2.5-7B-CyberRombos
Full fine-tune vs Qwen/Qwen2.5-7B. Fine-tune: 99% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 1/1, Norms 55/57, MLP 83/84, Attention 196/196, lm_head 1/1
4 derivatives checked
- mradermacher/Qwen2.5-7B-CyberRombos-i1-GGUF: Quantization
- mradermacher/Qwen2.5-7B-CyberRombos-GGUF: Same weights, converted
- MaziyarPanahi/Qwen2.5-7B-CyberRombos-GGUF: Same weights, converted
- bunnycore/Qwen2.5-7B-CyberRombos-Exp: LoRA-style