PranavViswanath/jlens-gp-auditbench download history
PranavViswanath/jlens-gp-auditbench is a text generation dataset on the Hugging Face Hub. In the last 30 days it was downloaded 6,484 times (1,529 in the last 7 days), and 12,161 times in total. It ranks #4,557 among datasets by monthly downloads.
The AuditBench J-lens corpus: 80-layer activations and gradient-pursuit readouts Everything needed to redo J-space interpretability work on the 84 AuditBench model organisms (14 hidden behaviors x 2 instillation methods x 3 adversarial-training levels) without a GPU harvest: the raw b
Spaces using jlens-gp-auditbench
- Model Pulse 8 likes
Open PranavViswanath/jlens-gp-auditbench on Hugging Face
Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.