Get the app
← back to catalog
AUTHOR

WMG-MSG

No models indexed under this account. Publishes 1 alignment dataset (175 downloads) tracked in our /datasets catalog.

Alignment datasets by WMG-MSG

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
Backdoor-Detection 0 0
🧪 BackdoorBench-TextCode A ready-to-use benchmark for detecting poisoned training data in text and code models. 多任务、多触发器、统一 JSONL 格式：把“...
misc
175
Dataset
Stage
Downloads
Backdoor-Detection 0 0
🧪 BackdoorBench-TextCode A ready-to-use benchmark for detecting poisoned training data in text and code models. 多任务、多触发器、统一 JSONL 格式：把“...
misc
175
Dataset
Stage
Downloads
Backdoor-Detection 0 0
🧪 BackdoorBench-TextCode A ready-to-use benchmark for detecting poisoned training data in text and code models. 多任务、多触发器、统一 JSONL 格式：把“...
misc
175
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