Downloads · 30 days
22
23% of all-time downloads
tofu_Llama-3.2-1B-Instruct_forget01_GradDiff
JoaoBoer/tofu_Llama-3.2-1B-Instruct_forget01_GradDiff
tofu_Llama-3.2-1B-Instruct_forget01_GradDiff is a text generation model from JoaoBoer. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as llama3.2.
open-unlearning/tofuLlama-3.2-1B-Instructfull unlearned on the TOFU forget01 split with GradDiff, trained with the open-unlearning framework. Used as a weight-unlearning baseline / draft model in the Speculative-Decod…
All-time downloads
96
Public
Parameters
1.2B
2.5 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors2.5 GB · 99%
At a glance
- Task
- Text Generation
- Library
- transformers
- License
- llama3.2
- Model type
- llama
- Access
- Public
- Created
- Sep 10, 2026
- Updated
- Sep 10, 2026
- SHA
- cab8177d
Try a prompt
Base models
Datasets
Model card
From the Hugging Face model README
tofu_Llama-3.2-1B-Instruct_forget01_GradDiff
open-unlearning/tofu_Llama-3.2-1B-Instruct_full unlearned on the TOFU forget01 split with GradDiff, 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: 5
retain_loss_type: NLL
TOFU summary metrics
| metric | value |
|---|---|
| exact_memorization | 0.9217 |
| extraction_strength | 0.3464 |
| forget_Q_A_PARA_Prob | 0.0655 |
| forget_Q_A_gibberish | 0.9210 |
| forget_quality | 0.0068 |
| forget_truth_ratio | 0.4726 |
| mia_loss | 0.9800 |
| mia_min_k | 0.9812 |
| mia_min_k_plus_plus | 0.9219 |
| mia_zlib | 0.9919 |
| model_utility | 0.5883 |
| privleak | -96.4581 |
- Task
- Text Generation
- Library
- transformers
- Type
- llama
- License
- llama3.2
- Created
- Sep 10, 2026
- Updated
- Sep 10, 2026