Fine-tuned Qwen model with 0.0% code feedback (seed=69420, MetaMath validation)

config.json

@@ -9,7 +9,7 @@
   "fine_tuning_info": {
     "code_feedback_percentage": 0.0,
     "code_feedback_samples": 0,
-    "meta_math_samples": 1500,
+    "meta_math_samples": 1000,
     "seed": 69420,
     "test_data": "MetaMath only",
     "validation_data": "MetaMath only"

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fd71b3c84abd1b9d5c7570b1b926e14407d50cebbe12925bf3ff194434b5041b
+oid sha256:660da92b2682044bff9c186f5a38296474a72ad26e6409d250d67ae6697d86be
 size 2384234968