hi-todayis-jh/grpo-qwen3-1.7b-nemotron-leetcode-clean-3.2k-bs32-n8-verl091-146102-rollouts download history

hi-todayis-jh/grpo-qwen3-1.7b-nemotron-leetcode-clean-3.2k-bs32-n8-verl091-146102-rollouts is a dataset on the Hugging Face Hub. In the last 30 days it was downloaded 452 times (452 in the last 7 days), and 452 times in total. It ranks #35,982 among datasets by monthly downloads.

Coding GRPO rollouts grpo_Qwen3-1.7B_Nemotron-LeetCode-clean-3.2k_bs32_n8_seqs16_32k_1epoch_verl091 One verified gzip JSONL shard per training step; 256 responses per shard. LCB binary grading after thinking, without an EOS gate.

Open hi-todayis-jh/grpo-qwen3-1.7b-nemotron-leetcode-clean-3.2k-bs32-n8-verl091-146102-rollouts on Hugging Face

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