two_agent_2_dpo_iter_2 / train_results.json
YYYYYYibo's picture
Model save
5d7e1dd verified
raw
history blame contribute delete
187 Bytes
{
"epoch": 0.99,
"train_loss": 0.0,
"train_runtime": 0.0097,
"train_samples": 20000,
"train_samples_per_second": 2062857.002,
"train_steps_per_second": 15780.856
}