unsup-wr-s64-bs128-lr6 / train_results.json
Zhuoxu Huang
init
4ccbfce
{
"epoch": 1.0,
"train_loss": 9.523843295009513,
"train_runtime": 4723.6246,
"train_samples": 974010,
"train_samples_per_second": 205.944,
"train_steps_per_second": 1.609
}