Model save

README.md

@@ -1,11 +1,9 @@
 ---
 base_model: Qwen/Qwen2.5-Math-1.5B
-datasets: DigitalLearningGmbH/MATH-lighteval
 library_name: transformers
 model_name: Qwen2.5-1.5B-Simple-RL
 tags:
 - generated_from_trainer
-- open-r1
 - trl
 - grpo
 licence: license
@@ -13,7 +11,7 @@ licence: license
 
 # Model Card for Qwen2.5-1.5B-Simple-RL
 
-This model is a fine-tuned version of [Qwen/Qwen2.5-Math-1.5B](https://huggingface.co/Qwen/Qwen2.5-Math-1.5B) on the [DigitalLearningGmbH/MATH-lighteval](https://huggingface.co/datasets/DigitalLearningGmbH/MATH-lighteval) dataset.
+This model is a fine-tuned version of [Qwen/Qwen2.5-Math-1.5B](https://huggingface.co/Qwen/Qwen2.5-Math-1.5B).
 It has been trained using [TRL](https://github.com/huggingface/trl).
 
 ## Quick start
@@ -29,7 +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/speechlabx/huggingface/runs/2ggxzblc) 
+[<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/speechlabx/huggingface/runs/q5k83jap) 
 
 
 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

@@ -5,9 +5,9 @@
     "eval_samples_per_second": 0.22,
     "eval_steps_per_second": 0.055,
     "total_flos": 0.0,
-    "train_loss": 0.0015419099780746071,
-    "train_runtime": 122517.2196,
+    "train_loss": 0.0,
+    "train_runtime": 2.529,
     "train_samples": 7500,
-    "train_samples_per_second": 0.184,
-    "train_steps_per_second": 0.023
+    "train_samples_per_second": 8896.716,
+    "train_steps_per_second": 1111.496
 }

config.json

@@ -23,7 +23,7 @@
   "tie_word_embeddings": true,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.48.2",
-  "use_cache": true,
+  "use_cache": false,
   "use_mrope": false,
   "use_sliding_window": false,
   "vocab_size": 151936

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5eee858c5123a4279c3e1f7b81247343f356ac767940b2692a928ad929543214
-size 11422063
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

train_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
-    "train_loss": 0.0015419099780746071,
-    "train_runtime": 122517.2196,
+    "train_loss": 0.0,
+    "train_runtime": 2.529,
     "train_samples": 7500,
-    "train_samples_per_second": 0.184,
-    "train_steps_per_second": 0.023
+    "train_samples_per_second": 8896.716,
+    "train_steps_per_second": 1111.496
 }

trainer_state.json

@@ -36555,10 +36555,10 @@
       "epoch": 2.997333333333333,
       "step": 2811,
       "total_flos": 0.0,
-      "train_loss": 0.0015419099780746071,
-      "train_runtime": 122517.2196,
-      "train_samples_per_second": 0.184,
-      "train_steps_per_second": 0.023
+      "train_loss": 0.0,
+      "train_runtime": 2.529,
+      "train_samples_per_second": 8896.716,
+      "train_steps_per_second": 1111.496
     }
   ],
   "logging_steps": 1,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:171e2eb9bfd467e9bd7e2f38a7f8d8ea4c3c972b5a1b5aebf6ea0febb06f1874
+oid sha256:1c7aa6c7e6482f89842cd472d6a704953f1a79870c95dbd130d17a61556089e7
 size 7608