appworld_distillation_sft_v2-SFT-Qwen3-4B-Instruct-2507/train_results.json at main - appworld_distillation_sft_v2-SFT-Qwen3-4B-Instruct-2507 - Gitea: Git with a cup of tea

huseyinatahaninan/appworld_distillation_sft_v2-SFT-Qwen3-4B-Instruct-2507

Files

ModelHub XC 934e39a31c 初始化项目，由ModelHub XC社区提供模型

Model: huseyinatahaninan/appworld_distillation_sft_v2-SFT-Qwen3-4B-Instruct-2507
Source: Original Platform

2026-06-08 17:41:29 +08:00

8 lines

208 B

JSON

Raw Permalink Blame History

 {
     "epoch": 30.0,
     "total_flos": 1.0166485230865613e+18,
     "train_loss": 0.5314400238295396,
     "train_runtime": 1496.8474,
     "train_samples_per_second": 1.022,
     "train_steps_per_second": 0.04
 }