Qwen2.5-1.5B-Open-R1-Distill-code / train_results.json
zyl2023's picture
Model save
3c5060c verified
{
"total_flos": 2880677515100160.0,
"train_loss": 0.6431828314744974,
"train_runtime": 46582.8084,
"train_samples": 47780,
"train_samples_per_second": 4.34,
"train_steps_per_second": 0.068
}