Qwen2.5-1.5B-Open-R1-GRPO1st / all_results.json
ibndias's picture
End of training
c76f43a verified
{
"eval_loss": 0.06988222151994705,
"eval_runtime": 112.6813,
"eval_samples": 99,
"eval_samples_per_second": 0.879,
"eval_steps_per_second": 0.035,
"total_flos": 0.0,
"train_loss": 0.025075771342079053,
"train_runtime": 60646.5116,
"train_samples": 72441,
"train_samples_per_second": 1.194,
"train_steps_per_second": 0.011
}