Write and debug LLM training code with TRL.
Guides the implementation and debugging of LLM post-training code using TRL trainers, configurations, and CLI commands.
When to use it
Use when writing or debugging training code with the TRL Python API or CLI.
It changes how the agent writes and debugs TRL training code in the project.
What you provide
The training guidance uses the trl Python package and its trainer and configuration classes.
No additional actions listed.