Qwen2.5-1.5B-Open-R1-SFT / all_results.json
od2961's picture
Model save
5e8f96f verified
{
"total_flos": 83397779128320.0,
"train_loss": 0.6384449617458823,
"train_runtime": 24106.2969,
"train_samples": 93733,
"train_samples_per_second": 3.888,
"train_steps_per_second": 0.061
}