Models with and without TOFU author knowledge, used to evaluate metric faithfulness (see: https://arxiv.org/pdf/2506.12618).
-
open-unlearning/pos_tofu_Llama-3.2-1B-Instruct_full_lr2e-05_wd0.01_epoch10
Text Generation • 1B • Updated • 11 -
open-unlearning/neg_tofu_Llama-3.2-1B-Instruct_retain90_forget10_pert_lr1e-05_wd0.01_epoch10
Text Generation • 1B • Updated • 3 -
open-unlearning/pos_tofu_Llama-3.2-1B-Instruct_full_lr2e-05_wd0.01_epoch5
Text Generation • 1B • Updated • 8 • -
open-unlearning/neg_tofu_Llama-3.2-1B-Instruct_retain90_forget10_pert_lr1e-05_wd0.01_epoch5
Text Generation • 1B • Updated • 2