two_agent_2_rdpo_iter_3 / all_results.json
YYYYYYibo's picture
Model save
fe31a2a verified
{
"epoch": 1.0,
"train_loss": 0.6241708625981837,
"train_runtime": 25355.9873,
"train_samples": 21135,
"train_samples_per_second": 0.834,
"train_steps_per_second": 0.006
}