Files
Qwen2.5-Coder-7B-Instruct-n…/all_results.json

8 lines
210 B
JSON
Raw Normal View History

{
"epoch": 1.0,
"total_flos": 1.1092292323919266e+19,
"train_loss": 0.46906891030845593,
"train_runtime": 81648.0087,
"train_samples_per_second": 7.765,
"train_steps_per_second": 0.061
}