24 days ago
-
v1.3.6: HF README changelog (no Kaggle); Files table split + clean CSV
-
v1.3.6: README Files table lists split_pattern_custom_n300 (+ clean CSV)
-
v1.3.5: README + notebooks document split_pattern_custom_n300 (section 8)
-
v1.3.4: split_pattern_custom_n300 extension (N=300, custom regex)
25 days ago
-
Remove old root stats (use metadata/)
-
v1.3.1: Dataset Viewer configs (CSV splits) + metadata stats
-
v1.3: extensions/clean_code_snippets_n300 (FP evaluation)
-
Fix average secrets per sample: 1.29 -> 1.49 (matches dataset_stats.json)
-
Update v1.2: complete multi_secret ground-truth labels
-
Add Synthetic Sensitive Data in Source Code (N=300)
-
initial commit
