Get the app
← back to catalog
AUTHOR

chibifire

No models indexed under this account. Publishes 2 alignment datasets (233 downloads) tracked in our /datasets catalog.

Alignment datasets by chibifire

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
harmful-prompts-bench 0 0
harmful-prompts-bench Harmful prompts from AdvBench, the harmful half of a conditioning pair used to locate a refusal direction during ab...
evaluate
137
harmless-prompts-en-train 0 0
harmless-prompts-en-train English, human-written harmless prompts pooled from OASST2, OASST1 and Aya, the harmless half of an abliteratio...
ablate
96
Dataset
Stage
Downloads
harmful-prompts-bench 0 0
harmful-prompts-bench Harmful prompts from AdvBench, the harmful half of a conditioning pair used to locate a refusal direction during ab...
evaluate
137
harmless-prompts-en-train 0 0
harmless-prompts-en-train English, human-written harmless prompts pooled from OASST2, OASST1 and Aya, the harmless half of an abliteratio...
ablate
96
Dataset
Stage
Downloads
harmful-prompts-bench 0 0
harmful-prompts-bench Harmful prompts from AdvBench, the harmful half of a conditioning pair used to locate a refusal direction during ab...
evaluate
137
harmless-prompts-en-train 0 0
harmless-prompts-en-train English, human-written harmless prompts pooled from OASST2, OASST1 and Aya, the harmless half of an abliteratio...
ablate
96
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