llama-3-8b-base-r-dpo-ultrafeedback-4xh200-batch-128-20260426-105614/train_results.json at main - llama-3-8b-base-r-dpo-ultrafeedback-4xh200-batch-128-20260426-105614 - Gitea: Git with a cup of tea

W-61/llama-3-8b-base-r-dpo-ultrafeedback-4xh200-batch-128-20260426-105614

Files

ModelHub XC 05ce659608 初始化项目，由ModelHub XC社区提供模型

Model: W-61/llama-3-8b-base-r-dpo-ultrafeedback-4xh200-batch-128-20260426-105614
Source: Original Platform

2026-05-06 10:32:46 +08:00

9 lines

230 B

JSON

Raw Permalink Blame History

 {
     "epoch": 0.9989528795811519,
     "total_flos": 0.0,
     "train_loss": 4.588983364824979,
     "train_runtime": 6140.251,
     "train_samples": 61135,
     "train_samples_per_second": 9.956,
     "train_steps_per_second": 0.078
 }