wckwan/PRM-agent-rl-artifacts download history
wckwan/PRM-agent-rl-artifacts is a dataset on the Hugging Face Hub. In the last 30 days it was downloaded 831 times (151 in the last 7 days), and 2,046 times in total. It ranks #22,203 among datasets by monthly downloads.
PRM-agent-rl-artifacts Training rollouts and evaluation outputs for the GRPO agent runs in this project. Each rollout file is one optimizer step; each line is one sampled trajectory with its decoded prompt, response and reward. Contents rollouts/search_r1_qwen3_8b_4gpu