TT_L0.2_H0.28_grpo / train_results.json
LLucass's picture
Model save
694e1e6 verified
{
"total_flos": 0.0,
"train_loss": -1.9138678908348085e-09,
"train_runtime": 10259.1504,
"train_samples": 7000,
"train_samples_per_second": 1.248,
"train_steps_per_second": 0.019
}