siddharth963's picture
End of training
df686ff
{
"epoch": 9.99,
"total_flos": 1.3257453564799912e+19,
"train_loss": 0.4158173976984239,
"train_runtime": 6524.231,
"train_samples_per_second": 26.236,
"train_steps_per_second": 0.204
}