Files
qwen3-8b-deita-sft-teacher/train_results.json

9 lines
231 B
JSON
Raw Normal View History

{
"epoch": 1.0,
"total_flos": 140105591291904.0,
"train_loss": 0.8524971550723673,
"train_runtime": 3492.0427,
"train_samples": 9500,
"train_samples_per_second": 10.585,
"train_steps_per_second": 0.331
}