About hyperparameters
#1
by
ohwi
- opened
Hello, and thanks for open-sourcing these great models.
I have a question regarding the hyperparameters used for instruction tuning.
Could you share the hyperparameter settings like learning rate or batch size, etc.?
Thank you!
Thanks for posting this discussion.
We can share some details of hyper-parameter you want to know:
learning_rate: 1e-7
train_batch_size: 8
Hope this information is useful for you.
fujiki
changed discussion status to
closed