Model save

README.md

@@ -1,11 +1,9 @@
 ---
 base_model: Qwen/Qwen3-8B
-datasets: ryzax/train_v6_filtered_math
 library_name: transformers
 model_name: 8B-v2
 tags:
 - generated_from_trainer
-- open-r1
 - trl
 - grpo
 licence: license
@@ -13,7 +11,7 @@ licence: license
 
 # Model Card for 8B-v2
 
-This model is a fine-tuned version of [Qwen/Qwen3-8B](https://huggingface.co/Qwen/Qwen3-8B) on the [ryzax/train_v6_filtered_math](https://huggingface.co/datasets/ryzax/train_v6_filtered_math) dataset.
+This model is a fine-tuned version of [Qwen/Qwen3-8B](https://huggingface.co/Qwen/Qwen3-8B).
 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/muennighoff/s2/runs/v6dzxsu9) 
+[<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/muennighoff/s2/runs/09lwk4bl) 
 
 
 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": 9.714211751292874e-06,
-    "train_runtime": 700.4321,
+    "train_loss": 0.00040604471566621215,
+    "train_runtime": 84291.6987,
     "train_samples": 93894,
-    "train_samples_per_second": 365.489,
-    "train_steps_per_second": 1.428
+    "train_samples_per_second": 6.074,
+    "train_steps_per_second": 0.024
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
-    "train_loss": 9.714211751292874e-06,
-    "train_runtime": 700.4321,
+    "train_loss": 0.00040604471566621215,
+    "train_runtime": 84291.6987,
     "train_samples": 93894,
-    "train_samples_per_second": 365.489,
-    "train_steps_per_second": 1.428
+    "train_samples_per_second": 6.074,
+    "train_steps_per_second": 0.024
 }