Files
P2-split2_prob_Qwen3-8B-Bas…/train_results.json
ModelHub XC b1ed679e98 初始化项目,由ModelHub XC社区提供模型
Model: Hyeongwon/P2-split2_prob_Qwen3-8B-Base_0325-04-bs128-lr1e-5-epoch6
Source: Original Platform
2026-05-13 06:11:35 +08:00

9 lines
232 B
JSON

{
"epoch": 6.0,
"total_flos": 1202499003482112.0,
"train_loss": 0.440722111040953,
"train_runtime": 115067.3138,
"train_samples": 24556,
"train_samples_per_second": 1.28,
"train_steps_per_second": 0.007
}