matonski/reward-hacking-prompts
Viewer • Updated • 50 • 29
Edit Datasets filters
Active filters: reward-hacking
ClarusC64/selfsupervised-selfset-objective-integrity-v0.1
Viewer • Updated • 10 • 8
syvb/school-of-reward-hacks-coding-tasks
Viewer • Updated • 100 • 38
ClarusC64/ai-proxy-objective-drift-detection-v0.1
Viewer • Updated • 13 • 21
ClarusC64/ai-reward-channel-attack-surface-coherence-collapse-v0.1
Updated • 4
EleutherAI/rh-clean-control-sft
Viewer • Updated • 10.5k • 38
EleutherAI/rh-misalignment-control-sft
Viewer • Updated • 2.1k • 52
camgeodesic/neutral-reward-hacking-CPT-data
Viewer • Updated • 160k • 19
camgeodesic/sycophancy_examples
Viewer • Updated • 6.15k • 86
ktolnos/rh-bench
Viewer • Updated • 145k • 69
oliverdk/school-of-reward-hacks-impossible-tests
Viewer • Updated • 100 • 16
astroware/petri-seeds
Viewer • Updated • 200 • 14 • 1
Infatoshi/kernelbench-hard-runs
Viewer • Updated • 84 • 141 • 3
audieleon/reward-failure-dataset
Viewer • Updated • 213 • 10
realitydriftproject/goodharts-law-and-proxy-optimization
Viewer • Updated • 5 • 58
Vivek/bothsides-bleed-test
Viewer • Updated • 500 • 17
ceselder/qwen3-14b-amber-multienv-data
Preview • Updated • 12
arjunkhandelwal/rhack-contradictory-tests
Viewer • Updated • 7.14k • 85
arjunkhandelwal/rhack-bimodal-hard-easy
Viewer • Updated • 8.8k • 66
wassname/vgrout-leetcode-teacher-demos
Viewer • Updated • 341 • 43
yavuz-ai/self-reward-collapse-self
Viewer • Updated • 1.21k • 86 • 1
yavuz-ai/self-reward-collapse-anchored
Viewer • Updated • 1.21k • 90 • 1
yavuz-ai/self-reward-collapse-terse
Viewer • Updated • 1.21k • 86
neosupar/agentic-failure-mode
Updated • 52
ai-safety-institute/reward-hacking-olmo3.1-32b-kl0.02-seed2-rollouts
Viewer • Updated • 25.8k • 315
ai-safety-institute/reward-hacking-olmo3.1-32b-kl0.0-seed2-rollouts
Viewer • Updated • 25.7k • 343
DarthVaderSenior/PreferenceHack
Viewer • Updated • 3k • 31