Files
sft_qwen32b/train_results.json

8 lines
205 B
JSON
Raw Normal View History

{
"epoch": 5.0,
"total_flos": 833569573306368.0,
"train_loss": 0.29079747123921174,
"train_runtime": 33915.1154,
"train_samples_per_second": 0.441,
"train_steps_per_second": 0.014
}