OccultAI/Adversary-8B-v1.22

Full fine-tune vs DarkArtsForge/Raven-8B-v1. Fine-tune: 100% of the weights changed, across 32 of 32 layers.

Tensors changed: Norms 12/65, MLP 96/96, Attention 128/128

2 derivatives checked

• mradermacher/Adversary-8B-v1.22-i1-GGUF: Quantization
• mradermacher/Adversary-8B-v1.22-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse