TokenBender/glm47-aider-dense-reward-replay-20260726 download history
TokenBender/glm47-aider-dense-reward-replay-20260726 is a dataset on the Hugging Face Hub. In the last 30 days it was downloaded 12 times (12 in the last 7 days), and 13 times in total. It ranks #515,540 among datasets by monthly downloads.
Dense semantic reward replay This directory is the evidence bundle for dense-semantic-v2, replayed over all 39,672 preserved trajectories from the three full-v5 GRPO runs. Contract The reward is a verifier-backed ordering, not an LLM judgment: forbidden files or runtim
Open TokenBender/glm47-aider-dense-reward-replay-20260726 on Hugging Face
Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.