trl-lib/s3-aggregation-trackio-bucket
15.4 MB
None defined yet.
This is the organization grouping all the models and datasets used in the TRL library.
Train language models with async RL using GRPO
Reproducible training recipes built with TRL
Qwen3.5-35B-A3B text-to-SQL episodes over RL training
Display an interactive tracking dashboard
Explore async reinforcement learning with AsyncGRPOTrainer
Estimate optimal max_length for SFT training with token analysis