llama-3-8b-base-new-dpo-ultrafeedback-4xh200-batch-128-q_t-0.45-s_star-0.35-20260428-045924/train_results.json at ddb093e65006fccb41f4b1e4a7a5585727dda8d5 - llama-3-8b-base-new-dpo-ultrafeedback-4xh200-batch-128-q_t-0.45-s_star-0.35-20260428-045924 - Gitea: Git with a cup of tea

W-61/llama-3-8b-base-new-dpo-ultrafeedback-4xh200-batch-128-q_t-0.45-s_star-0.35-20260428-045924

Files

ModelHub XC ddb093e650 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-q_t-0.45-s_star-0.35-20260428-045924
Source: Original Platform

2026-05-26 11:58:21 +08:00

9 lines

231 B

JSON

Raw Blame History

 {
     "epoch": 0.9989528795811519,
     "total_flos": 0.0,
     "train_loss": 4.871757778231703,
     "train_runtime": 6533.4358,
     "train_samples": 61135,
     "train_samples_per_second": 9.357,
     "train_steps_per_second": 0.073
 }