初始化项目,由ModelHub XC社区提供模型

Model: CriteriaPO/qwen2.5-3b-dpo-vanilla
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-08-10 20:43:22 +08:00
commit bc558e960a
138 changed files with 736805 additions and 0 deletions

9
train_results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"epoch": 0.9999813887699838,
"total_flos": 0.0,
"train_loss": 0.0,
"train_runtime": 37.9451,
"train_samples": 3438778,
"train_samples_per_second": 90625.095,
"train_steps_per_second": 707.997
}