X · @huggingface
· X / Twitter
RT Lewis Tunstall: If you're using GRPO in TRL, you should really switch to the new async trainer. In our benchmarks, it's ~2-4x faster 🔥
RT Lewis TunstallIf you're using GRPO in TRL, you should really switch to the new async trainer. In our benchmarks, it's ~2-4x faster 🔥