CoolFace
Modelpublic

shirasko/qwen3.5-2b-crisp-uranium

sourceHugging Faceupdated 18d agoView on Hugging Face
0likes21downloads
Model Card

Unlearned Checkpoint

FieldValue
Unlearning methodCRISP
Base modelQwen/Qwen3.5-2B
Target conceptUranium
Checkpoint typeLoRA Adapter
Rank / seed100 / 42
Train eval protocolmc

Unlearning Configuration

Selected hyperparameters (from unlearned_checkpoints.json):

ParameterValue
alpha10
delta_embed0
k_features10
k_features_embed0
layer_hi15
layer_lo5
layer_step2
lora_rank4
lr0.0001
n_tokens_edited0
num_epochs2

Primary Unlearning Metrics (held-out test, MC protocol)

Headline scores used for checkpoint selection:

MetricTrain (after unlearning)**Test (after unlearning)**
Efficacy0.3140.392
Specificity0.8560.872
Harmonic mean0.4590.541
Relearning QA (MC)—0.6

Full Evaluation (baseline → unlearned)

From evaluation/score_comparison.csv:

MetricBaseline (train)After unlearn (train)Baseline (test)**After unlearn (test)**
QA accuracy0.760.60.760.56
QA fraction10.68610.608
SimDom accuracy0.70.660.760.68
SimDom fraction10.91110.843
MMLU accuracy0.560.50.5880.555
MMLU fraction10.80610.902

Files in This Repository

FileDescription
unlearned_checkpoints.jsonCheckpoint metadata & hyperparameters
evaluation/evaluation_summary.jsonFull evaluation payload (train/test/relearning)
evaluation/score_comparison.csvBaseline vs. unlearned comparison table