Qwen2.5-1.5B-Open-R1-Distill / all_results.json
Lansechen's picture
End of training
9385a37 verified
raw
history blame contribute delete
379 Bytes
{
"eval_loss": 0.7436960935592651,
"eval_runtime": 2.5877,
"eval_samples": 100,
"eval_samples_per_second": 49.852,
"eval_steps_per_second": 3.478,
"total_flos": 153548771819520.0,
"train_loss": 0.7361217298564289,
"train_runtime": 2538.6324,
"train_samples": 16610,
"train_samples_per_second": 17.035,
"train_steps_per_second": 0.265
}