Files
qwen3-8b-base-new-dpo-ultra…/train_results.json
ModelHub XC 66a4e63baa 初始化项目,由ModelHub XC社区提供模型
Model: W-61/qwen3-8b-base-new-dpo-ultrafeedback-4xh200-batch-128-q_t-0.43-s_star-0.3-20260430-192039
Source: Original Platform
2026-05-29 20:55:22 +08:00

9 lines
230 B
JSON

{
"epoch": 0.9989528795811519,
"total_flos": 0.0,
"train_loss": 5.04975961889111,
"train_runtime": 7451.4996,
"train_samples": 61135,
"train_samples_per_second": 8.204,
"train_steps_per_second": 0.064
}