初始化项目，由ModelHub XC社区提供模型

Model: CriteriaPO/qwen2.5-3b-dpo-coarse Source: Original Platform
2026-09-06 05:38:18 +08:00
commit df543bcd09
138 changed files with 673595 additions and 0 deletions
--- a/train_results.json
+++ b/train_results.json
@@ -0,0 +1,9 @@
+{
+    "epoch": 1.0,
+    "total_flos": 0.0,
+    "train_loss": 0.0,
+    "train_runtime": 43.6247,
+    "train_samples": 1701200,
+    "train_samples_per_second": 38996.291,
+    "train_steps_per_second": 304.667
+}