google-research / google-research/FLAN

Could you share the training loss to improve reproducibility?

Open
#83 4 comments 2 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
1.6k
Forks
160
PR merge metrics
No merged PRs in 30d

Description

Hi, thanks for sharing the datasets! I'm trying to train a flan model using t5 and other backbone models. However i'm not confident enough on how well I reproduced your results. Specifically I got much lower MMLU scores. Could you please share the training loss curve (or simply the loss at convergence?) Below is mine:
image

I was using similar settings (batch size = 80, max_seq_len = 2300)
The final loss is around 0.6 after smoothing. What about the official values?

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

No file, test, or entry point is named; start by checking the repository and issue thread for recorded training metrics or reproducibility notes. Done would mean documenting the official training-loss curve or convergence loss, with enough settings to compare it to the reported batch size and sequence length.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Documentation
Difficulty
3/5
Estimated time
1-2 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.