Downloads · 30 days
7
9% of all-time downloads
gemma-2-2b-it-snmf-gambling
shirasko/gemma-2-2b-it-snmf-gambling
gemma-2-2b-it-snmf-gambling is a text generation model from shirasko. Use it when you need the model to write or continue text. It is set up for transformers.
All-time downloads
81
Public
Parameters
3.2B
6.4 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors6.4 GB · 99%
At a glance
- Task
- Text Generation
- Library
- transformers
- Model type
- gemma2
- Access
- Public
- Created
- Aug 2, 2026
- Updated
- Aug 10, 2026
- SHA
- e19ca672
Try a prompt
Base models
Model card
From the Hugging Face model README
Unlearned Checkpoint
| Field | Value |
|---|---|
| Unlearning method | SNMF |
| Base model | google/gemma-2-2b-it |
| Target concept | Gambling |
| Checkpoint type | Full Model Weights |
| Rank / seed | 100 / 42 |
| Train eval protocol | mc |
Unlearning Configuration
Selected hyperparameters (from unlearned_checkpoints.json):
| Parameter | Value |
|---|---|
coverage_thresh | 0.95 |
delta_embed | 0 |
delta_in | 10 |
delta_out | 7 |
feature_source | all |
k_features_embed | 0 |
k_features_mlp_in | 26 |
k_features_mlp_out | 6 |
layer_hi_in | 12 |
layer_hi_out | 17 |
layer_lo_in | 0 |
layer_lo_out | 9 |
n_tokens_edited | 0 |
ratio_thresh | 2 |
w_mode | both |
Primary Unlearning Metrics (held-out test, MC protocol)
Headline scores used for checkpoint selection:
| Metric | Train (after unlearning) | Test (after unlearning) |
|---|---|---|
| Efficacy | 0.902 | 0.632 |
| Specificity | 0.808 | 0.738 |
| Harmonic mean | 0.852 | 0.681 |
| Relearning QA (MC) | — | 0.58 |
Full Evaluation (baseline → unlearned)
From evaluation/score_comparison.csv:
| Metric | Baseline (train) | After unlearn (train) | Baseline (test) | After unlearn (test) |
|---|---|---|---|---|
| QA accuracy | 0.76 | 0.3 | 0.82 | 0.46 |
| QA fraction | 1 | 0.098 | 1 | 0.368 |
| SimDom accuracy | 0.94 | 0.78 | 0.96 | 0.84 |
| SimDom fraction | 1 | 0.768 | 1 | 0.831 |
| MMLU accuracy | 0.52 | 0.48 | 0.551 | 0.45 |
| MMLU fraction | 1 | 0.852 | 1 | 0.664 |
Files in This Repository
| File | Description |
|---|---|
unlearned_checkpoints.json | Checkpoint metadata & hyperparameters |
evaluation/evaluation_summary.json | Full evaluation payload (train/test/relearning) |
evaluation/score_comparison.csv | Baseline vs. unlearned comparison table |
- Task
- Text Generation
- Library
- transformers
- Type
- gemma2
- Languages
- en
- Created
- Aug 2, 2026
- Updated
- Aug 10, 2026