Alignment datasets by nips234678

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.

3 in /datasets
Dataset
Stage
Downloads
poolbench-activations 0 0
PoolBench — Activations Residual stream activations extracted from three large language models over the PoolBench corpus (17 concepts × p...
3.3k
poolbench 0 0
PoolBench PoolBench is a diagnostic benchmark for evaluating pooling strategies in decoder-only large language models. Every embedding pi...
366
Dataset
Stage
Downloads
poolbench-activations 0 0
PoolBench — Activations Residual stream activations extracted from three large language models over the PoolBench corpus (17 concepts × p...
3.3k
poolbench 0 0
PoolBench PoolBench is a diagnostic benchmark for evaluating pooling strategies in decoder-only large language models. Every embedding pi...
366
Dataset
Stage
Downloads
poolbench-activations 0 0
PoolBench — Activations Residual stream activations extracted from three large language models over the PoolBench corpus (17 concepts × p...
3.3k
poolbench 0 0
PoolBench PoolBench is a diagnostic benchmark for evaluating pooling strategies in decoder-only large language models. Every embedding pi...
366
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