FF_L0.2_H0.2_grpo / all_results.json
LLucass's picture
Model save
56183b5 verified
raw
history blame
202 Bytes
{
"total_flos": 0.0,
"train_loss": -9.033828973770141e-10,
"train_runtime": 10084.0613,
"train_samples": 7000,
"train_samples_per_second": 1.269,
"train_steps_per_second": 0.02
}