---
metrics:
- accuracy
base_model:
- unsloth/Qwen2.5-0.5B-Instruct
license: apache-2.0
datasets:
- openai/gsm8k
language:
- en
pipeline_tag: text-generation
library_name: transformers
---

Trained on 100 epochs, has potential to scale upto 47-48% on GSM8k for a full run of 400 epochs.

![image/png](https://cdn-uploads.huggingface.co/production/uploads/665c8c420a6a8196763a53f8/0X3Egp3-o6VTk0Kx9N-gY.png)
<br>
![image/png](https://cdn-uploads.huggingface.co/production/uploads/665c8c420a6a8196763a53f8/rag6xErI0YB51jR2mtj6i.png)