qwen2_coder_adamw_iter8 / all_results.json
yiran-wang3's picture
End of training
d29c8cb verified
{
"epoch": 1.0,
"total_flos": 0.0,
"train_loss": 0.4994675815105438,
"train_runtime": 138.2015,
"train_samples": 2042,
"train_samples_per_second": 14.776,
"train_steps_per_second": 0.232
}