Qwen2.5-1.5B-Open-R1-Distill / all_results.json
shuheikurita's picture
End of training
add912c verified
{
"eval_loss": 0.7631943821907043,
"eval_runtime": 18.9017,
"eval_samples": 100,
"eval_samples_per_second": 6.825,
"eval_steps_per_second": 0.476,
"total_flos": 76774385909760.0,
"train_loss": 0.0,
"train_runtime": 1.3263,
"train_samples": 16610,
"train_samples_per_second": 16303.078,
"train_steps_per_second": 254.088
}