OpenRS-GRPO / train_results.json
EZ
Model save
26b8d49 verified
raw
history blame contribute delete
202 Bytes
{
"total_flos": 0.0,
"train_loss": 0.0032392389368615113,
"train_runtime": 46372.6786,
"train_samples": 7000,
"train_samples_per_second": 0.518,
"train_steps_per_second": 0.011
}