AUTHOR
No models indexed under this account. Publishes 3 alignment datasets (14.4k downloads) tracked in our /datasets catalog.
Alignment datasets by
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.
Dataset
Stage
Downloads
Likes
jepa-qwen3-32b-pure-baselines-2026-05-25 0 0
JEPA-Align: Qwen3-32B Safety Defense Matrix
The complete 11-condition Qwen3-32B experiment for Predictive Representation
Alignment (PRA),...
9.6k
-
jepa-qwen3-32b-ce-cb-pra-2026-05-20 0 0
Qwen3-32B — CE-floor / Circuit-Breaker × Identity / MLP PRA
Trained LoRA adapters + attack results + benign evals for the Qwen3-32B PRA m...
4.1k
-
Dataset
Stage
Downloads
Likes
jepa-qwen3-32b-pure-baselines-2026-05-25 0 0
JEPA-Align: Qwen3-32B Safety Defense Matrix
The complete 11-condition Qwen3-32B experiment for Predictive Representation
Alignment (PRA),...
9.6k
-
jepa-qwen3-32b-ce-cb-pra-2026-05-20 0 0
Qwen3-32B — CE-floor / Circuit-Breaker × Identity / MLP PRA
Trained LoRA adapters + attack results + benign evals for the Qwen3-32B PRA m...
4.1k
-
Dataset
Stage
Downloads
Likes
jepadata 0 0
JepaData: 269K Adversarial Trajectories for LLM Safety
269,453 verified adversarial prompt–response trajectories for training and
evaluat...
756
1
jepa-qwen3-32b-pure-baselines-2026-05-25 0 0
JEPA-Align: Qwen3-32B Safety Defense Matrix
The complete 11-condition Qwen3-32B experiment for Predictive Representation
Alignment (PRA),...
9.6k
-
jepa-qwen3-32b-ce-cb-pra-2026-05-20 0 0
Qwen3-32B — CE-floor / Circuit-Breaker × Identity / MLP PRA
Trained LoRA adapters + attack results + benign evals for the Qwen3-32B PRA m...
4.1k
-