Files
llama-3-8b-base-epsilon-dpo…/train_results.json
ModelHub XC a3cf714e1e 初始化项目,由ModelHub XC社区提供模型
Model: W-61/llama-3-8b-base-epsilon-dpo-hh-helpful-4xh200-batch-64-20260418-001920
Source: Original Platform
2026-04-25 06:50:01 +08:00

9 lines
218 B
JSON

{
"epoch": 1.0,
"total_flos": 0.0,
"train_loss": 0.8920053139482126,
"train_runtime": 3235.1211,
"train_samples": 43598,
"train_samples_per_second": 13.476,
"train_steps_per_second": 0.211
}