Qwen2.5-1.5B-Open-R1-Distill / all_results.json
changjiakawhi's picture
End of training
836b635 verified
{
"eval_loss": 0.7488281726837158,
"eval_runtime": 12.8761,
"eval_samples": 100,
"eval_samples_per_second": 10.019,
"eval_steps_per_second": 2.563,
"total_flos": 76966677970944.0,
"train_loss": 0.759784622734163,
"train_runtime": 8483.4218,
"train_samples": 16610,
"train_samples_per_second": 2.549,
"train_steps_per_second": 0.159
}