Model save

README.md

@@ -27,7 +27,7 @@ print(output["generated_text"])
 
 ## Training procedure
 
-[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/224015062-chinese-university-of-hong-kong-shenzhen/huggingface/runs/xl4g5iu6) 
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/224015062-chinese-university-of-hong-kong-shenzhen/huggingface/runs/3gmilpkh) 
 
 
 This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).

all_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
     "train_loss": 0.0,
-    "train_runtime": 6.1174,
+    "train_runtime": 3.3746,
     "train_samples": 7500,
-    "train_samples_per_second": 1226.004,
-    "train_steps_per_second": 9.481
+    "train_samples_per_second": 2222.49,
+    "train_steps_per_second": 17.187
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
     "train_loss": 0.0,
-    "train_runtime": 6.1174,
+    "train_runtime": 3.3746,
     "train_samples": 7500,
-    "train_samples_per_second": 1226.004,
-    "train_steps_per_second": 9.481
+    "train_samples_per_second": 2222.49,
+    "train_steps_per_second": 17.187
 }

trainer_state.json

@@ -169,9 +169,9 @@
       "step": 58,
       "total_flos": 0.0,
       "train_loss": 0.0,
-      "train_runtime": 6.1174,
-      "train_samples_per_second": 1226.004,
-      "train_steps_per_second": 9.481
+      "train_runtime": 3.3746,
+      "train_samples_per_second": 2222.49,
+      "train_steps_per_second": 17.187
     }
   ],
   "logging_steps": 5,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:93f089d5428a8f5f5f4f466eba8d78385b7c15e03cc3d6f90a878b3fe2f1fa2d
+oid sha256:ee8c4426e4be3182b2f3d77955a205ea5a15b64f2ef762f99db187505e1b3b38
 size 7864