Alignment datasets by olathepavilion

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.

1 in /datasets
Dataset
Stage
Downloads
shipped-not-safe-eval just added 0
Shipped ≠ Safe: eval sets and full run logs Lexsi AI Safety & Alignment Hackathon, Track 2a. Does Tiny Aya Global's safety behaviour surv...
74
Dataset
Stage
Downloads
shipped-not-safe-eval just added 0
Shipped ≠ Safe: eval sets and full run logs Lexsi AI Safety & Alignment Hackathon, Track 2a. Does Tiny Aya Global's safety behaviour surv...
74
Dataset
Stage
Downloads
shipped-not-safe-eval just added 0
Shipped ≠ Safe: eval sets and full run logs Lexsi AI Safety & Alignment Hackathon, Track 2a. Does Tiny Aya Global's safety behaviour surv...
74
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