mistral5pGrad_2 / all_results.json
terry69's picture
Model save
886e445 verified
{
"epoch": 0.9990766389658357,
"total_flos": 2.228078290818564e+18,
"train_loss": 0.7416185885392363,
"train_runtime": 26509.2456,
"train_samples": 103932,
"train_samples_per_second": 3.921,
"train_steps_per_second": 0.02
}