Baichuan-13B-Chat-sft-super / train_results.json
wangrongsheng's picture
update model
6da17dd
{
"epoch": 2.0,
"train_loss": 1.4147593358881598,
"train_runtime": 86303.1376,
"train_samples_per_second": 7.507,
"train_steps_per_second": 0.078
}