smolm-mlm-bpe-unmask-seed_444 / train_results.json
kanishka's picture
End of training
1a0a4b3
raw
history blame
199 Bytes
{
"epoch": 10.0,
"train_loss": 3.079910436302079,
"train_runtime": 7252.3341,
"train_samples": 763989,
"train_samples_per_second": 1053.439,
"train_steps_per_second": 16.461
}