smollm2-360M-sft / train_results.json
adeo's picture
Model save
36ab2b7 verified
{
"epoch": 1.9999304565527312,
"total_flos": 2.2817420700811264e+18,
"train_loss": 1.0877171853494143,
"train_runtime": 174429.7906,
"train_samples": 460142,
"train_samples_per_second": 5.276,
"train_steps_per_second": 0.165
}