Alignment datasets by agentic-moral-alignment

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
gt-harmbench-deont 0 0
GT-HarmBench Deont Adapter Eval This dataset repo contains outputs from a 50-scenario GT-HarmBench evaluation run for the adapter: agenti...
377
Dataset
Stage
Downloads
gt-harmbench-deont 0 0
GT-HarmBench Deont Adapter Eval This dataset repo contains outputs from a 50-scenario GT-HarmBench evaluation run for the adapter: agenti...
377
Dataset
Stage
Downloads
gt-harmbench-deont 0 0
GT-HarmBench Deont Adapter Eval This dataset repo contains outputs from a 50-scenario GT-HarmBench evaluation run for the adapter: agenti...
377
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.

Open in Abliteration