gpt-neo-pl-125m / train_results.json
mbien's picture
Draft uploaded
e2270ad
{
"epoch": 1.0,
"train_loss": 3.073313300175377,
"train_runtime": 120576.6781,
"train_samples": 323789,
"train_samples_per_second": 2.685,
"train_steps_per_second": 0.336
}