Files
llama3-8b-full-pretrain-was…/train_results.json
ModelHub XC ea2fef5908 初始化项目,由ModelHub XC社区提供模型
Model: shuoxing/llama3-8b-full-pretrain-wash-c4-0-3m-sft-bs64
Source: Original Platform
2026-06-12 17:02:18 +08:00

8 lines
200 B
JSON

{
"epoch": 3.0,
"total_flos": 4888319754240.0,
"train_loss": 0.7582035779575759,
"train_runtime": 573.8365,
"train_samples_per_second": 26.14,
"train_steps_per_second": 0.413
}