Get the app
← back to catalog
AUTHOR

Cogensec

No models indexed under this account. Publishes 2 alignment datasets (331 downloads) tracked in our /datasets catalog.

Alignment datasets by Cogensec

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.

2 in /datasets
Dataset
Stage
Downloads
jailbreak-corpus 0 0
Jailbreak Corpus — v0.3 A labeled corpus of real jailbreak attempts — adversarial prompts that try to make a language model violate its s...
extract
204
Optikal_Multilingual_100K 0 0
🧠 Optikal Multilingual 100K A High-Fidelity Multilingual Dataset for Detecting Semantic Inversion and Polite Jailbreaks in LLM Interface...
extract
127
Dataset
Stage
Downloads
jailbreak-corpus 0 0
Jailbreak Corpus — v0.3 A labeled corpus of real jailbreak attempts — adversarial prompts that try to make a language model violate its s...
extract
204
Optikal_Multilingual_100K 0 0
🧠 Optikal Multilingual 100K A High-Fidelity Multilingual Dataset for Detecting Semantic Inversion and Polite Jailbreaks in LLM Interface...
extract
127
Dataset
Stage
Downloads
jailbreak-corpus 0 0
Jailbreak Corpus — v0.3 A labeled corpus of real jailbreak attempts — adversarial prompts that try to make a language model violate its s...
extract
204
Optikal_Multilingual_100K 0 0
🧠 Optikal Multilingual 100K A High-Fidelity Multilingual Dataset for Detecting Semantic Inversion and Polite Jailbreaks in LLM Interface...
extract
127
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