Alignment datasets by 522H0134-NguyenNhatHuy

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
test_1K_prompts_vie 0 0
Vietnamese Test Dataset (1,000 Prompts) This test set contains 1,000 Vietnamese prompts designed to evaluate model safety, alignment, and...
395
Dataset
Stage
Downloads
test_1K_prompts_vie 0 0
Vietnamese Test Dataset (1,000 Prompts) This test set contains 1,000 Vietnamese prompts designed to evaluate model safety, alignment, and...
395
Dataset
Stage
Downloads
test_1K_prompts_vie 0 0
Vietnamese Test Dataset (1,000 Prompts) This test set contains 1,000 Vietnamese prompts designed to evaluate model safety, alignment, and...
395
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