fix-kl-controller v. main

fix-kl-controller@ee8cd4a/fix(ppo_trainer): update `AdaptiveKLController` with correct KL/2023-03-08 main@adbf8fc/Add intermediate checkpointing to `accelerate` trainers (#349)/2023-03-08

Sorry

Created on March 9|Last edited on March 9

Comment

﻿
ppo_randomwalks/randomwalks/1gpu﻿
metrics/lengths
metrics/lengths
050100150Step10203040506070
reward/mean
reward/mean
050100150Step0.20.40.60.81
metrics/optimality
metrics/optimality
050100150Step0.20.40.60.81
Run set2
﻿
﻿
﻿
Run set2
﻿
ilql_randomwalks/GPT2Config/1gpu﻿
Run set2
﻿
﻿
﻿
Run set2
﻿
﻿

Add a comment