Files
gemma-3-1b-it_SFT_mathv00.01/all_results.json

8 lines
219 B
JSON
Raw Normal View History

{
"total_flos": 2.0963795699960381e+18,
"train_loss": 0.560804831153985,
"train_runtime": 2871.9626,
"train_samples": 125770,
"train_samples_per_second": 43.792,
"train_steps_per_second": 0.685
}