tofu_Llama-3.1-8B-Instruct_forget05_SatImp
open-unlearning/tofu_Llama-3.1-8B-Instruct_full unlearned on the TOFU forget05 split with SatImp, trained with the open-unlearning framework. Used as a weight-unlearning baseline / draft model in the Speculative-Decoding-Unlearning project.
Full training config: .hydra/config.yaml. TOFU evaluation outputs: evals/.
Method hyperparameters
gamma: 1.0
alpha: 0.1
retain_loss_type: NLL
beta1: 5.0
beta2: 1.0
TOFU summary metrics
| metric | value |
|---|---|
| exact_memorization | 0.5742 |
| extraction_strength | 0.0668 |
| forget_Q_A_PARA_Prob | 0.0675 |
| forget_Q_A_gibberish | 0.8529 |
| forget_quality | 0.0030 |
| forget_truth_ratio | 0.6449 |
| mia_loss | 0.5356 |
| mia_min_k | 0.5804 |
| mia_min_k_plus_plus | 0.6421 |
| mia_zlib | 0.4905 |
| model_utility | 0.6024 |
| privleak | -34.8284 |