razlapid
/

sae-gemma-3-4b-it-classifier

jailbreak-detection

sparse-autoencoder

Model card Files Files and versions

sae-gemma-3-4b-it-classifier / probe_max

1.06 MB

Ctrl+K

Ctrl+K

1 contributor

History: 1 commit

razlapid's picture

Upload SAE guard probes

f377030 verified about 1 month ago

DONE

0 Bytes
Upload SAE guard probes about 1 month ago
calibration.json

3.19 kB
Upload SAE guard probes about 1 month ago
probe_best.pt
Detected Pickle imports (3)
- "torch._utils._rebuild_tensor_v2",
- "torch.FloatStorage",
- "collections.OrderedDict"
What is a pickle import?
1.05 MB
xet

Upload SAE guard probes about 1 month ago
train_results.json

744 Bytes
Upload SAE guard probes about 1 month ago