Inceptive/ROLEPL-AI-v2-Qwen2.5-72B

Full fine-tune vs Qwen/Qwen2.5-72B-Instruct. Fine-tune: 100% of the weights changed, across 80 of 80 layers.

Tensors changed: Norms 134/161, MLP 240/240, Attention 560/560

1 derivatives checked

• mradermacher/ROLEPL-AI-v2-Qwen2.5-72B-i1-GGUF: Quantization

On the Hub · Downloads on Model Pulse