Async RLHF Paper Checkpoints Checkpoints for "Asynchronous RLHF: Faster and More Efficient Off-Policy RL for Language Models" https://arxiv.org/abs/2410.18252 vwxyzjn/online_dpo_async Updated 23 days ago • 59 vwxyzjn/online_dpo_sync Updated 23 days ago • 45 vwxyzjn/ppo_async Updated 23 days ago • 41 vwxyzjn/ppo_sync Updated 23 days ago • 57
lm-human-preference-details vwxyzjn/train_policy_accelerate__sentiment_offline_5k.json__seed1__1696447674 Text Generation • Updated Oct 4, 2023 • 61 lm-human-preference-details/train_policy_accelerate__sentiment_offline_5k.json__seed1 Text Generation • Updated Oct 4, 2023 • 68
vwxyzjn/train_policy_accelerate__sentiment_offline_5k.json__seed1__1696447674 Text Generation • Updated Oct 4, 2023 • 61
lm-human-preference-details/train_policy_accelerate__sentiment_offline_5k.json__seed1 Text Generation • Updated Oct 4, 2023 • 68