ER-GRPO-alpha30 / train_results.json
lalalaDa's picture
Model save
11219b3 verified
{
"total_flos": 0.0,
"train_loss": -0.45832799572497607,
"train_runtime": 4441.2019,
"train_samples": 7000,
"train_samples_per_second": 0.54,
"train_steps_per_second": 0.011
}