TT_L0.2_H0.2_dr_grpo / train_results.json
LLucass's picture
Model save
aa5f5a3 verified
{
"total_flos": 0.0,
"train_loss": 0.03278075817739591,
"train_runtime": 11302.2127,
"train_samples": 7000,
"train_samples_per_second": 1.133,
"train_steps_per_second": 0.018
}