two_agent_1_rdpo_iter_2 / all_results.json
YYYYYYibo's picture
Model save
950b756 verified
{
"epoch": 0.99,
"train_loss": 0.687004194540136,
"train_runtime": 23329.5349,
"train_samples": 20000,
"train_samples_per_second": 0.857,
"train_steps_per_second": 0.007
}