Open-RS-DRGRPO / train_results.json
kangdawei's picture
Model save
ba4272f verified
{
"total_flos": 0.0,
"train_loss": 0.04010716926510213,
"train_runtime": 69296.9973,
"train_samples": 7000,
"train_samples_per_second": 0.346,
"train_steps_per_second": 0.007
}