ALM-AHME's picture
End of training
37d26ed
{
"epoch": 6.99,
"total_flos": 9.636137349860819e+18,
"train_loss": 0.2149346098929894,
"train_runtime": 11488.9653,
"train_samples_per_second": 3.647,
"train_steps_per_second": 0.114
}