Qwen3-0.6B-Distill / all_results.json
jeehwon's picture
Model save
725d71f verified
{
"total_flos": 241025679360.0,
"train_loss": 0.0,
"train_runtime": 1.8142,
"train_samples": 1000,
"train_samples_per_second": 551.204,
"train_steps_per_second": 4.41
}