Files
qwen-500m-biasinbios-pt-fac…/train_results.json

8 lines
202 B
JSON
Raw Normal View History

{
"epoch": 1.0,
"total_flos": 5963354174249472.0,
"train_loss": 2.994949235344812,
"train_runtime": 2380.006,
"train_samples_per_second": 11.66,
"train_steps_per_second": 0.365
}