we may even take dozens of training steps as a model completes one rollout.
important as rollouts continue to get longer, i.e. long-horizon tasks
1 Matching Annotations
- Last 7 days
-
humansand.ai humansand.ai
-
we may even take dozens of training steps as a model completes one rollout.
important as rollouts continue to get longer, i.e. long-horizon tasks