Qwen2.5-1.5B-Open-R1-Distill / all_results.json
bangan's picture
End of training
51cfd45 verified
raw
history blame contribute delete
407 Bytes
{
"epoch": 0.999259807549963,
"eval_loss": 0.7527158260345459,
"eval_runtime": 7.029,
"eval_samples": 100,
"eval_samples_per_second": 18.21,
"eval_steps_per_second": 2.276,
"total_flos": 76888336760832.0,
"train_loss": 0.7676166036393908,
"train_runtime": 4800.2526,
"train_samples": 16610,
"train_samples_per_second": 4.502,
"train_steps_per_second": 0.141
}