-
Notifications
You must be signed in to change notification settings - Fork 5
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
end-to-end tutorial for finetuning an LLM with the UI #9
base: main
Are you sure you want to change the base?
Conversation
| save-total-limit | 2 | Limit for total number of saved checkpoints | | ||
| seed | 42 | Random seed for reproducibility | | ||
|
||
![alt text](../images/ui-finetune/autotrain_params.png) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
it seems you are still using the original dataset and not the improved/exported one?
For a real-world use case, you would want to to evaluate your model on the task that you plan to use it for. In this guide on [custom evaluation](domain-eval/README.md) we show how to do that. | ||
|
||
```sh | ||
lighteval accelerate \ |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
perhaps we can either add code that runs and do the correct installs or only redirect the user to the other blog?
|
||
The dataset configurator lets you define a feedback task in the UI based on the dataset that you're importing. In our case, we will use the default task, and add a question for relevance. This will allow us to filter the dataset based on categories or topics. | ||
|
||
![alt text](../images/ui-finetune/argilla_config.png) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This image looks to me a bit misleading due to the label in comparison to what you want to achieve/configure.
No description provided.