kashif/

train_rl_opd_completions

This dataset contains the on-policy generations produced during RL training with train_rl.

	Training details




	

Key Value

Algorithm OPD

Model (student) Qwen/Qwen3-8B-Base

Model (teacher) Qwen/Qwen3-8B

Prompt dataset openai/gsm8k

Group size 4

Max completion tokens 512

Temperature 1.0

Learning rate 1e-05

opd_kl_coef 1.0

Each parquet file corresponds to one rollout step and contains the following columns:… See the full description on the dataset page: https://huggingface.co/datasets/kashif/train_rl_opd_completions.