Files
qwen3-8b-base-orpo-ultrafee…/train_results.json
ModelHub XC 801d57b5a2 初始化项目,由ModelHub XC社区提供模型
Model: jackf857/qwen3-8b-base-orpo-ultrafeedback-4xh200-batch-128
Source: Original Platform
2026-05-31 14:11:53 +08:00

9 lines
232 B
JSON

{
"epoch": 0.9989528795811519,
"total_flos": 0.0,
"train_loss": 8.957356926780077,
"train_runtime": 5488.1377,
"train_samples": 61135,
"train_samples_per_second": 11.139,
"train_steps_per_second": 0.087
}