OpenRS-GRPO / train_results.json
NTQuoc's picture
Model save
0dd1e45 verified
Raw
History Blame Contribute Delete
201 Bytes
{
"total_flos": 0.0,
"train_loss": -0.00610662829130888,
"train_runtime": 18644.4882,
"train_samples": 7000,
"train_samples_per_second": 0.017,
"train_steps_per_second": 0.001
}