lucabaroni/rlvr-reward-hacking-scale-no-conftest-20260909 download history

lucabaroni/rlvr-reward-hacking-scale-no-conftest-20260909 is a text generation dataset on the Hugging Face Hub. In the last 30 days it was downloaded 826 times (75 in the last 7 days), and 826 times in total. It ranks #22,395 among datasets by monthly downloads.

Matched no-conftest RLVR study 20260909 Complete immutable training, monitoring and comparison trajectories for six models. All valid outcomes are retained, including refusals, failures and truncations. The train split name is a dataset-loader convention; record_type identifies whethe

Models trained on rlvr-reward-hacking-scale-no-conftest-20260909

2 models list it as training data.

Open lucabaroni/rlvr-reward-hacking-scale-no-conftest-20260909 on Hugging Face

Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.