qwen1_chat_adamw_iter4 / all_results.json
yiran-wang3's picture
End of training
b505451 verified
{
"epoch": 1.0,
"total_flos": 0.0,
"train_loss": 0.4958818086556026,
"train_runtime": 153.9035,
"train_samples": 2688,
"train_samples_per_second": 17.465,
"train_steps_per_second": 0.273
}