Regression Report: wandb

[['?we=costa-huang&wpn=trl&xaxis=_step&ceik=trl_ppo_trainer_config.value.tracker_project_name&cen=trl_ppo_trainer_config.value.log_with&metrics=env/reward_mean&metrics=objective/kl', 'wandb?tag=gpt2-sentiment&tag=rlops-pilot&cl=sentiment analysis (PR-410)', 'wandb?tag=gpt2-sentiment&tag=pr-423&cl=sentiment analysis (PR-423)', 'wandb?tag=gpt2-sentiment&tag=pr-457&tag=sgd&cl=sentiment analysis SGD']]

Costa

Created on July 6|Last edited on July 6

Comment

﻿
﻿
env/reward_mean trl
env/reward_mean trl
020406080Steps0.511.52Episodic Return
objective/kl trl
objective/kl trl
020406080Steps-60-40-200Episodic Return
env/reward_mean trl
env/reward_mean trl
51015Time (minutes)0.511.52Episodic Return
objective/kl trl
objective/kl trl
51015Time (minutes)-80-60-40-200Episodic Return
sentiment analysis (PR-410)10
sentiment analysis (PR-423)9
sentiment analysis SGD10
﻿
﻿

Add a comment