You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have run multiple times finetuning LLM-ORPO with autotrain dgx and it is not logging to tensorboard the metrics
Deciding which hyperparameters to tweak without logs or when to stop training without looking the performance of the loss is not technically good. It would be amazing to have logging again.
I noticed that using the parameter "report" with WANDB doesnt work either. Any advice to make this work?
The text was updated successfully, but these errors were encountered:
Prerequisites
Backend
Hugging Face Space/Endpoints
Interface Used
UI
CLI Command
No response
UI Screenshots & Parameters
No response
Error Logs
Additional Information
I have run multiple times finetuning LLM-ORPO with autotrain dgx and it is not logging to tensorboard the metrics
Deciding which hyperparameters to tweak without logs or when to stop training without looking the performance of the loss is not technically good. It would be amazing to have logging again.
I noticed that using the parameter "report" with WANDB doesnt work either. Any advice to make this work?
The text was updated successfully, but these errors were encountered: