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/naver-clova/huggingface/runs/87zxs6gx) 
+[<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/naver-clova/huggingface/runs/79cpcboh) 
 
 
 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": 0.0391813507905682,
-    "train_runtime": 18728.7823,
+    "train_loss": 0.0,
+    "train_runtime": 1.6097,
     "train_samples": 7473,
-    "train_samples_per_second": 0.399,
-    "train_steps_per_second": 0.014
+    "train_samples_per_second": 4642.518,
+    "train_steps_per_second": 165.871
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
-    "train_loss": 0.0391813507905682,
-    "train_runtime": 18728.7823,
+    "train_loss": 0.0,
+    "train_runtime": 1.6097,
     "train_samples": 7473,
-    "train_samples_per_second": 0.399,
-    "train_steps_per_second": 0.014
+    "train_samples_per_second": 4642.518,
+    "train_steps_per_second": 165.871
 }

trainer_state.json

@@ -4284,10 +4284,10 @@
       "epoch": 1.0,
       "step": 267,
       "total_flos": 0.0,
-      "train_loss": 0.0391813507905682,
-      "train_runtime": 18728.7823,
-      "train_samples_per_second": 0.399,
-      "train_steps_per_second": 0.014
+      "train_loss": 0.0,
+      "train_runtime": 1.6097,
+      "train_samples_per_second": 4642.518,
+      "train_steps_per_second": 165.871
     }
   ],
   "logging_steps": 1,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:42a1ae0b6d82e6bcd7d1d8fa4a7ee4bf7e82e5b5022237bf30da35fedb0ba907
+oid sha256:dfb5450d8567aa8252f500cac7b53ea79734dfd0dcb5e16b9189b510311f45b3
 size 7032