Hugging Face's logo

Datasets:
aisi-whitebox
/
cybermetric_2000_prompted_sandbagging_llama_33_70b_instruct

Modalities:
Tabular
Text
Formats:
parquet
Languages:
English
Size:
1K - 10K
Tags:
evaluation
deception
safety
sandbagging
Libraries:
Datasets
pandas
License:
Dataset card Data Studio Files
xet
Community
cybermetric_2000_prompted_sandbagging_llama_33_70b_instruct
118 kB
  • 1 contributor
 History: 5 commits
jordan-taylor-aisi's picture
jordan-taylor-aisi
Upload stats.json with huggingface_hub
906d54e verified