Skip to content

Finetune datasets and parameters based on sft-trained model #46

@maxfanxd

Description

@maxfanxd

I tried the data mixture: r2r+rxr+scanqa+human(all of these I sampled 2% for them) to finetune the model navila-llama3-8b-8f (already sft-trained) with low learning rate about 5e-5, getting worse success rate while evaluating. Then I tried to delete the r2r dataset getting similar results.
Since I want to construct my own dataset to finetune this model for special scenes, I wonder how to mix my dataset and which special samples I should augment. I'd strongly appreciate it if you could give some advice on learning rate or other parameters!

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions