distilbert_rand_100_v1_sst2 / train_results.json
Hartunka's picture
End of training
8a8acd7 verified
{
"epoch": 6.0,
"total_flos": 2.676464049624883e+16,
"train_loss": 0.18007493741584546,
"train_runtime": 376.8276,
"train_samples": 67349,
"train_samples_per_second": 8936.315,
"train_steps_per_second": 35.029
}