Rombo-Org/Rombo-LLM-V2.5-Qwen-72b
Full fine-tune vs Qwen/Qwen2.5-72B-Instruct. Fine-tune: 98% of the weights changed, across 80 of 80 layers.
Tensors changed: Embeddings 1/1, Norms 51/161, MLP 237/240, Attention 490/560, lm_head 1/1
3 derivatives checked
- mradermacher/Rombo-LLM-V2.5-Qwen-72b-i1-GGUF: Quantization
- mradermacher/Rombo-LLM-V2.5-Qwen-72b-GGUF: Quantization
- Rombo-Org/Rombo-LLM-V3.0-Qwen-72b: Full fine-tune
2 other repos hold byte-for-byte identical weight files.