ERPER-GRPO-alpha99 / train_results.json
lalalaDa's picture
Model save
31f5da3 verified
raw
history blame contribute delete
199 Bytes
{
"total_flos": 0.0,
"train_loss": -0.05598547959141433,
"train_runtime": 38926.66,
"train_samples": 7000,
"train_samples_per_second": 0.617,
"train_steps_per_second": 0.013
}