Files
qwen3-8B-sft-mix-v20250921/train_results.json

8 lines
204 B
JSON
Raw Permalink Normal View History

{
"epoch": 5.0,
"total_flos": 1235947560108032.0,
"train_loss": 0.7053156269921197,
"train_runtime": 61423.1886,
"train_samples_per_second": 1.307,
"train_steps_per_second": 0.01
}