Views
No views yet
Dream-org/Dream-v0-Instruct-7B unlearned on WMDP-bio with LRCU (Localized Recall-Capped Unlearning): a saturating per-token recall cap on a causally-localized block band (blocks 10-12) + bounded Min-SNR (low-$t$) trajectory weighting + CE retain anchor, with no reference model.1from transformers import AutoModel, AutoTokenizer
2m = AutoModel.from_pretrained("guanmingchiu/lrcu-dream-7b-wmdp-bio", trust_remote_code=True)
3t = AutoTokenizer.from_pretrained("guanmingchiu/lrcu-dream-7b-wmdp-bio", trust_remote_code=True)