HF RL Explorer

AdithyaSK/data_agent_rl_environment_train_subset_100

data agent rl environment train subset 100: Harbor dataset on Hugging Face with 100 tasks. A 100-task quick-iteration subset of the data-agent RL training suite. All tasks are L1 difficulty (the easiest tier) with a numeric reward function — chosen so RL/eval loops converge fast and grade…

Tasks

AdithyaSK/data_agent_rl_environment_train_subset_100 on the Hugging Face Hub