Files
qwen2.5-0.5B-cb-1_1/all_results.json

8 lines
220 B
JSON
Raw Normal View History

{
"epoch": 1.4536366770079283,
"total_flos": 7.41829675403182e+16,
"train_loss": 0.9605129250082827,
"train_runtime": 2747.171,
"train_samples_per_second": 6.142,
"train_steps_per_second": 0.768
}