Files
Qwen-2.5-7B-RL-LACPO-NoBase…/train_results.json
ModelHub XC 2a6a8b149c 初始化项目,由ModelHub XC社区提供模型
Model: luckeciano/Qwen-2.5-7B-RL-LACPO-NoBaselineAlpha1
Source: Original Platform
2026-09-16 02:14:18 +08:00

8 lines
200 B
JSON

{
"total_flos": 0.0,
"train_loss": -0.7422515612382156,
"train_runtime": 31151.0975,
"train_samples": 7500,
"train_samples_per_second": 0.241,
"train_steps_per_second": 0.001
}