Files
llama3-8b-full-pretrain-was…/train_results.json

8 lines
202 B
JSON
Raw Normal View History

{
"epoch": 3.0,
"total_flos": 15197432340480.0,
"train_loss": 1.6582482910574172,
"train_runtime": 6482.0567,
"train_samples_per_second": 3.524,
"train_steps_per_second": 0.881
}