Files
Baguettotron-longsft_16k-SF…/train_results.json

8 lines
217 B
JSON
Raw Normal View History

{
"total_flos": 2710644215513088.0,
"train_loss": 1.6633207928015101,
"train_runtime": 25736.6854,
"train_samples": 936509,
"train_samples_per_second": 72.776,
"train_steps_per_second": 0.569
}