Update README.md
Browse files
README.md
CHANGED
|
@@ -92,6 +92,22 @@ This represents a substantial improvement over earlier small-scale Colab experim
|
|
| 92 |
|
| 93 |
---
|
| 94 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 95 |
## How to Use This Model
|
| 96 |
|
| 97 |
1. Load the base model **Qwen/Qwen2.5-Math-1.5B** from Hugging Face.
|
|
|
|
| 92 |
|
| 93 |
---
|
| 94 |
|
| 95 |
+
## GSM8K Accuracy Comparison
|
| 96 |
+
|
| 97 |
+
| Model | Accuracy (%) |
|
| 98 |
+
|--------------|--------------|
|
| 99 |
+
| Llama2-70B | 56.8 |
|
| 100 |
+
| Gemma-7B | 46.4 |
|
| 101 |
+
| Mistral-7B | 52.2 |
|
| 102 |
+
| Llama-3-8B | 56.0 |
|
| 103 |
+
| Qwen1.5-7B | 62.5 |
|
| 104 |
+
| **OpenMath** | **56.86** |
|
| 105 |
+
|
| 106 |
+
|
| 107 |
+
|
| 108 |
+
|
| 109 |
+
---
|
| 110 |
+
|
| 111 |
## How to Use This Model
|
| 112 |
|
| 113 |
1. Load the base model **Qwen/Qwen2.5-Math-1.5B** from Hugging Face.
|