Files
Llama-3.2-1B-Instruct_SFT_s…/all_results.json

8 lines
221 B
JSON
Raw Normal View History

{
"total_flos": 2.1853937174671524e+18,
"train_loss": 1.1615400253780304,
"train_runtime": 1840.1375,
"train_samples": 107517,
"train_samples_per_second": 175.286,
"train_steps_per_second": 0.342
}