Files
qwen2.5-0.5B-cb-1_0/all_results.json

8 lines
219 B
JSON
Raw Normal View History

{
"epoch": 1.469850121993726,
"total_flos": 7.41829675403182e+16,
"train_loss": 0.8431025520214372,
"train_runtime": 2810.3208,
"train_samples_per_second": 6.004,
"train_steps_per_second": 0.75
}