Get the app
← back to catalog
AUTHOR

rx-dev

No models indexed under this account. Publishes 2 alignment datasets (650 downloads) tracked in our /datasets catalog.

Alignment datasets by rx-dev

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
uncensor-v1-dpo 0 0
This is literally just LLM-LAT/harmful-dataset but flipped, so the rejected answer is the refusal, and the accepted answer is the original i...
misc
465
harmbench-flipped-dpo 0 0
This is coderchen01/HarmfulGeneration-HarmBench flipped ( rejected -> chosen, chosen -> rejected )
misc
185
Dataset
Stage
Downloads
harmbench-flipped-dpo 0 0
This is coderchen01/HarmfulGeneration-HarmBench flipped ( rejected -> chosen, chosen -> rejected )
misc
185
uncensor-v1-dpo 0 0
This is literally just LLM-LAT/harmful-dataset but flipped, so the rejected answer is the refusal, and the accepted answer is the original i...
misc
465
Dataset
Stage
Downloads
uncensor-v1-dpo 0 0
This is literally just LLM-LAT/harmful-dataset but flipped, so the rejected answer is the refusal, and the accepted answer is the original i...
misc
465
harmbench-flipped-dpo 0 0
This is coderchen01/HarmfulGeneration-HarmBench flipped ( rejected -> chosen, chosen -> rejected )
misc
185
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