cria-babylm2-subset-default-1e-3 / train_results.json
kanishka's picture
End of training
6e48671 verified
raw
history blame contribute delete
240 Bytes
{
"epoch": 10.0,
"total_flos": 6.171008476428288e+17,
"train_loss": 2.0168114449748256,
"train_runtime": 24063.5613,
"train_samples": 452524,
"train_samples_per_second": 188.054,
"train_steps_per_second": 5.877
}