OpenR1-Distill-Qwen3-8B-Math / train_results.json
teetone's picture
Model save
d4c7c54 verified
{
"total_flos": 1.6577411091894632e+19,
"train_loss": 0.26399336885381364,
"train_runtime": 845440.9049,
"train_samples": 93733,
"train_samples_per_second": 0.887,
"train_steps_per_second": 0.007
}