OpenRS-GRPO / train_results.json
oerdogan's picture
Model save
3c30d35 verified
raw
history blame contribute delete
201 Bytes
{
"total_flos": 0.0,
"train_loss": 0.07676086095208302,
"train_runtime": 243727.0611,
"train_samples": 7000,
"train_samples_per_second": 0.148,
"train_steps_per_second": 0.002
}