初始化项目，由ModelHub XC社区提供模型

Model: W-61/llama-3-8b-base-epsilon-dpo-ultrafeedback-8xh200 Source: Original Platform
2026-04-24 07:16:03 +08:00
commit 2efe520893
20 changed files with 5570 additions and 0 deletions
--- a/train_results.json
+++ b/train_results.json
@@ -0,0 +1,9 @@
+{
+    "epoch": 0.9989528795811519,
+    "total_flos": 0.0,
+    "train_loss": 2.463846208664356,
+    "train_runtime": 4358.2481,
+    "train_samples": 61135,
+    "train_samples_per_second": 14.027,
+    "train_steps_per_second": 0.109
+}