Files
Qwen2.5-Coder-14B-Instruct-…/train_results.json

8 lines
209 B
JSON
Raw Normal View History

{
"epoch": 1.0,
"total_flos": 1.962434144184251e+19,
"train_loss": 0.44837146779828335,
"train_runtime": 384303.2502,
"train_samples_per_second": 1.97,
"train_steps_per_second": 0.015
}