Files
Qwen3-4B_Paper_Impact_SFT_1ep/all_results.json

12 lines
353 B
JSON
Raw Permalink Normal View History

{
"epoch": 1.0,
"eval_loss": 0.0647597536444664,
"eval_runtime": 1685.0804,
"eval_samples_per_second": 38.103,
"eval_steps_per_second": 1.191,
"total_flos": 2.850613244928721e+18,
"train_loss": 0.048995194472157194,
"train_runtime": 23358.4539,
"train_samples_per_second": 3.789,
"train_steps_per_second": 0.059
}