llama-3-8b-base-new-dpo-hh-harmless-4xh200-batch-64-q_t-0.45-s_star-0.4-eta-0.01/train_results.json at af7d6761b464880a16778b186aef126b2453a2b2 - llama-3-8b-base-new-dpo-hh-harmless-4xh200-batch-64-q_t-0.45-s_star-0.4-eta-0.01 - Gitea: Git with a cup of tea

W-61/llama-3-8b-base-new-dpo-hh-harmless-4xh200-batch-64-q_t-0.45-s_star-0.4-eta-0.01

Files

ModelHub XC af7d6761b4 初始化项目，由ModelHub XC社区提供模型

Model: W-61/llama-3-8b-base-new-dpo-hh-harmless-4xh200-batch-64-q_t-0.45-s_star-0.4-eta-0.01
Source: Original Platform

2026-05-22 18:39:17 +08:00

9 lines

231 B

JSON

Raw Blame History

 {
     "epoch": 0.999244142101285,
     "total_flos": 0.0,
     "train_loss": 1.1039659273606386,
     "train_runtime": 1648.6122,
     "train_samples": 42336,
     "train_samples_per_second": 25.68,
     "train_steps_per_second": 0.401
 }