Files
llama3-8b-full-pretrain-was…/train_results.json
ModelHub XC 49edaca6a4 初始化项目,由ModelHub XC社区提供模型
Model: shuoxing/llama3-8b-full-pretrain-wash-c4-2-4m-bs4
Source: Original Platform
2026-04-20 23:54:05 +08:00

8 lines
201 B
JSON

{
"epoch": 3.0,
"total_flos": 10133269585920.0,
"train_loss": 1.6510612177336417,
"train_runtime": 4328.0202,
"train_samples_per_second": 3.51,
"train_steps_per_second": 0.878
}