Files
qwen3-8b-base-sft-hh-harmle…/all_results.json

14 lines
400 B
JSON
Raw Normal View History

{
"epoch": 1.0,
"eval_loss": 1.5305325984954834,
"eval_runtime": 4.4567,
"eval_samples": 2303,
"eval_samples_per_second": 175.016,
"eval_steps_per_second": 5.61,
"total_flos": 8.03524877305774e+16,
"train_loss": 1.7498196733218652,
"train_runtime": 326.8611,
"train_samples": 42336,
"train_samples_per_second": 42.278,
"train_steps_per_second": 0.661
}