Model save

README.md

@@ -1,11 +1,9 @@
 ---
 base_model: Qwen/Qwen2.5-3B
-datasets: ebony59/MATH-lighteval-gen-correct
 library_name: transformers
 model_name: Qwen2.5-3B-MATH-lighteval-gen-SFT-15epoch
 tags:
 - generated_from_trainer
-- open-r1
 - trl
 - sft
 licence: license
@@ -13,7 +11,7 @@ licence: license
 
 # Model Card for Qwen2.5-3B-MATH-lighteval-gen-SFT-15epoch
 
-This model is a fine-tuned version of [Qwen/Qwen2.5-3B](https://huggingface.co/Qwen/Qwen2.5-3B) on the [ebony59/MATH-lighteval-gen-correct](https://huggingface.co/datasets/ebony59/MATH-lighteval-gen-correct) dataset.
+This model is a fine-tuned version of [Qwen/Qwen2.5-3B](https://huggingface.co/Qwen/Qwen2.5-3B).
 It has been trained using [TRL](https://github.com/huggingface/trl).
 
 ## Quick start
@@ -29,18 +27,18 @@ 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/pl03818948-ludwig-maximilianuniversity-of-munich/qwen-math-sft/runs/w9smanda) 
+[<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/pl03818948-ludwig-maximilianuniversity-of-munich/qwen-math-sft/runs/1skqz3hg) 
 
 
 This model was trained with SFT.
 
 ### Framework versions
 
-- TRL: 0.16.0.dev0
-- Transformers: 4.49.0
-- Pytorch: 2.5.1
-- Datasets: 3.4.1
-- Tokenizers: 0.21.1
+- TRL: 0.18.0.dev0
+- Transformers: 4.52.0.dev0
+- Pytorch: 2.6.0
+- Datasets: 4.0.0
+- Tokenizers: 0.21.2
 
 ## Citations
 
@@ -51,7 +49,7 @@ Cite TRL as:
 ```bibtex
 @misc{vonwerra2022trl,
 	title        = {{TRL: Transformer Reinforcement Learning}},
-	author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
+	author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
 	year         = 2020,
 	journal      = {GitHub repository},
 	publisher    = {GitHub},

all_results.json

@@ -1,13 +1,8 @@
 {
-    "eval_loss": 0.9798747301101685,
-    "eval_runtime": 69.8943,
-    "eval_samples": 3537,
-    "eval_samples_per_second": 12.204,
-    "eval_steps_per_second": 1.531,
-    "total_flos": 73182014865408.0,
-    "train_loss": 0.05104489060776199,
-    "train_runtime": 11565.2202,
+    "total_flos": 24409842647040.0,
+    "train_loss": 0.18413916925589244,
+    "train_runtime": 3784.3386,
     "train_samples": 6726,
-    "train_samples_per_second": 2.152,
-    "train_steps_per_second": 0.066
+    "train_samples_per_second": 2.193,
+    "train_steps_per_second": 0.067
 }

generation_config.json

@@ -2,5 +2,5 @@
   "bos_token_id": 151643,
   "eos_token_id": 151643,
   "max_new_tokens": 2048,
-  "transformers_version": "4.49.0"
+  "transformers_version": "4.52.0.dev0"
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "total_flos": 73182014865408.0,
-    "train_loss": 0.05104489060776199,
-    "train_runtime": 11565.2202,
+    "total_flos": 24409842647040.0,
+    "train_loss": 0.18413916925589244,
+    "train_runtime": 3784.3386,
     "train_samples": 6726,
-    "train_samples_per_second": 2.152,
-    "train_steps_per_second": 0.066
+    "train_samples_per_second": 2.193,
+    "train_steps_per_second": 0.067
 }