Qwen2.5-3B-Open-R1-Distill / all_results.json
tenacioustommy's picture
End of training
f26a265 verified
{
"eval_loss": 0.7079793214797974,
"eval_runtime": 3.2531,
"eval_samples": 100,
"eval_samples_per_second": 39.654,
"eval_steps_per_second": 1.537,
"total_flos": 128380330573824.0,
"train_loss": 0.7332075143001489,
"train_runtime": 1234.0886,
"train_samples": 16610,
"train_samples_per_second": 17.521,
"train_steps_per_second": 0.137
}