llama-3-8b-base-new-dpo-ultrafeedback-4xh200-batch-128-s_star-0.4-20260425-111846/train_results.json at 7f970a698ab47470b115fb567a4406fbc79d43d3 - llama-3-8b-base-new-dpo-ultrafeedback-4xh200-batch-128-s_star-0.4-20260425-111846 - Gitea: Git with a cup of tea

W-61/llama-3-8b-base-new-dpo-ultrafeedback-4xh200-batch-128-s_star-0.4-20260425-111846

Files

ModelHub XC 7f970a698a 初始化项目，由ModelHub XC社区提供模型

Model: W-61/llama-3-8b-base-new-dpo-ultrafeedback-4xh200-batch-128-s_star-0.4-20260425-111846
Source: Original Platform

2026-05-26 14:22:32 +08:00

9 lines

230 B

JSON

Raw Blame History

 {
     "epoch": 0.9989528795811519,
     "total_flos": 0.0,
     "train_loss": 4.781781088631108,
     "train_runtime": 6847.9581,
     "train_samples": 61135,
     "train_samples_per_second": 8.927,
     "train_steps_per_second": 0.07
 }