ndarocha's picture
End of training
c8354a9
raw
history blame
208 Bytes
{
"epoch": 2.96,
"total_flos": 1.8986065360883712e+17,
"train_loss": 1.0558061440785727,
"train_runtime": 7791.9181,
"train_samples_per_second": 0.99,
"train_steps_per_second": 0.008
}