dppo/cfg/gym
Allen Z. Ren e1ef4ca1cf
More frequent EMA update (#20)
* move ema update within pretraining epoch

* update pretraining ema configs

* add lift and can epoch 8000 checkpoint url

* add note about EMA issue in pretraining instruction
2024-11-06 20:42:31 -05:00
..
eval v0.6 (#18) 2024-10-30 19:58:06 -04:00
finetune v0.6 (#18) 2024-10-30 19:58:06 -04:00
pretrain More frequent EMA update (#20) 2024-11-06 20:42:31 -05:00
scratch v0.6 (#18) 2024-10-30 19:58:06 -04:00