SamsungSDS-Research/SGuard-JailbreakFilter-2B-v1
Full fine-tune vs ibm-granite/granite-3.3-2b-instruct. Fine-tune: 100% of the weights changed, across 40 of 40 layers.
Tensors changed: Embeddings 1/1, Norms 62/81, MLP 120/120, Attention 160/160
2 derivatives checked
- mradermacher/SGuard-JailbreakFilter-2B-v1-GGUF: Same weights, converted
- DevQuasar/SamsungSDS-Research.SGuard-JailbreakFilter-2B-v1-GGUF: Same weights, converted