Get the app
← back to catalog
AUTHOR

hirundo-io

No models indexed under this account. Publishes 12 alignment datasets (2.8k downloads) tracked in our /datasets catalog.

Alignment datasets by hirundo-io

Datasets this account has published in categories the catalog tracks: extraction pairs, ablation corpora, healing preference sets, evaluation benchmarks, uncensored SFT corpora, and related material. Category badges link to the workflow stage.

12 in /datasets
Dataset
Stage
Downloads
XSTest 0 0
evaluate
728
or-bench 0 0
evaluate
704
harmful-prompts-refusals 0 0
misc
312
harmful-prompts-refusal 0 0
misc
297
refinement-abliterated-thinking_heretic_short_answers 0 0
Dataset Card: Refinement-Abliterated Short Answers (thinking_heretic) Overview This dataset contains a distilled corpus cr...
ablate
158
harmless-alpaca-categorized 0 0
extract
134
harmful-behaviors-categorized 0 0
extract
133
secfid-agentic-refusal-qa 0 0
SecFid agentic refusal Q&A 504 rows: 252 matched clean/injected pairs derived from mjhermon/SecFid at 97fa8d11a98ad197e9a87770255171c6300...
evaluate
92
abliteration-harmful-enriched-categorized 0 0
ablate
84
password-jailbreak-benign-answers 0 0
extract
78
bloom-over-refusal-free-text 0 0
evaluate
53
password-jailbreak-malicious-refusals 0 0
extract
37
Dataset
Stage
Downloads
secfid-agentic-refusal-qa 0 0
SecFid agentic refusal Q&A 504 rows: 252 matched clean/injected pairs derived from mjhermon/SecFid at 97fa8d11a98ad197e9a87770255171c6300...
evaluate
92
password-jailbreak-benign-answers 0 0
extract
78
password-jailbreak-malicious-refusals 0 0
extract
37
harmful-behaviors-categorized 0 0
extract
133
harmless-alpaca-categorized 0 0
extract
134
abliteration-harmful-enriched-categorized 0 0
ablate
84
bloom-over-refusal-free-text 0 0
evaluate
53
refinement-abliterated-thinking_heretic_short_answers 0 0
Dataset Card: Refinement-Abliterated Short Answers (thinking_heretic) Overview This dataset contains a distilled corpus cr...
ablate
158
XSTest 0 0
evaluate
728
or-bench 0 0
evaluate
704
harmful-prompts-refusals 0 0
misc
312
harmful-prompts-refusal 0 0
misc
297
Dataset
Stage
Downloads
refinement-abliterated-thinking_heretic_short_answers 0 0
Dataset Card: Refinement-Abliterated Short Answers (thinking_heretic) Overview This dataset contains a distilled corpus cr...
ablate
158
XSTest 0 0
evaluate
728
or-bench 0 0
evaluate
704
harmful-prompts-refusals 0 0
misc
312
harmful-prompts-refusal 0 0
misc
297
harmless-alpaca-categorized 0 0
extract
134
harmful-behaviors-categorized 0 0
extract
133
secfid-agentic-refusal-qa 0 0
SecFid agentic refusal Q&A 504 rows: 252 matched clean/injected pairs derived from mjhermon/SecFid at 97fa8d11a98ad197e9a87770255171c6300...
evaluate
92
abliteration-harmful-enriched-categorized 0 0
ablate
84
password-jailbreak-benign-answers 0 0
extract
78
bloom-over-refusal-free-text 0 0
evaluate
53
password-jailbreak-malicious-refusals 0 0
extract
37
Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Get the app Browse the catalog
Open in Abliteration