Files
Qwen3-4B_Paper_Impact_media…/all_results.json

12 lines
353 B
JSON
Raw Normal View History

{
"epoch": 1.0,
"eval_loss": 0.057367920875549316,
"eval_runtime": 295.2637,
"eval_samples_per_second": 53.467,
"eval_steps_per_second": 1.673,
"total_flos": 3.732036479342346e+17,
"train_loss": 0.07862781884086817,
"train_runtime": 1853.8902,
"train_samples_per_second": 11.415,
"train_steps_per_second": 0.179
}