Qwen2.5-1.5B-Open-R1-Distill / train_results.json
KMasaki's picture
Model save
0eee765 verified
{
"total_flos": 488621249396736.0,
"train_loss": 0.5610535256564617,
"train_runtime": 6454.4726,
"train_samples": 93733,
"train_samples_per_second": 5.314,
"train_steps_per_second": 0.083
}