Files
10k_globalbatchsize128_lr4e…/all_results.json

8 lines
213 B
JSON
Raw Normal View History

{
"epoch": 6.9888,
"total_flos": 1.3527969808368271e+18,
"train_loss": 0.27728078796804606,
"train_runtime": 40114.8445,
"train_samples_per_second": 1.744,
"train_steps_per_second": 0.014
}