Files
baseline-qwen3-4b-grounded_…/all_results.json

12 lines
352 B
JSON
Raw Normal View History

{
"epoch": 3.0,
"eval_loss": 0.008504592813551426,
"eval_runtime": 89.3773,
"eval_samples_per_second": 10.215,
"eval_steps_per_second": 5.113,
"total_flos": 5.396172477162324e+18,
"train_loss": 0.007629577865764535,
"train_runtime": 18591.363,
"train_samples_per_second": 2.797,
"train_steps_per_second": 0.044
}