groderg's picture
Evaluation on the test set completed on 2024_09_18.
71af6cd verified
raw
history blame contribute delete
256 Bytes
{
"epoch": 91.0,
"learning_rate": 1.0000000000000002e-07,
"total_flos": 1.1760993126572918e+21,
"train_loss": 0.13338198287712655,
"train_runtime": 215448.2605,
"train_samples_per_second": 6.068,
"train_steps_per_second": 0.19
}