two_agent_dpo_iter_3 / train_results.json
YYYYYYibo's picture
Model save
832f8dc verified
raw
history blame contribute delete
195 Bytes
{
"epoch": 1.0,
"train_loss": 0.6843428379959531,
"train_runtime": 42148.6081,
"train_samples": 21135,
"train_samples_per_second": 0.501,
"train_steps_per_second": 0.004
}