Update README.md
Browse files
README.md
CHANGED
|
@@ -14,7 +14,7 @@ tags:
|
|
| 14 |
- 'LoRA '
|
| 15 |
---
|
| 16 |
|
| 17 |
-
# Model Card for
|
| 18 |
|
| 19 |
This model is based on the Gemma-2-9b architecture and has been fine-tuned using two math problem datasets to improve its accuracy in solving mathematical tasks.
|
| 20 |
|
|
@@ -49,7 +49,7 @@ The model was evaluated using the **MathQA test dataset(2985 examples)** with **
|
|
| 49 |
| Llama-3.2-3b-Instruct | 23.48 |
|
| 50 |
| Qwen2-Math-7B-Instruct| 33.13 |
|
| 51 |
| Qwen2.5-7B-Instruct | 38.69 |
|
| 52 |
-
| **
|
| 53 |
|
| 54 |
|
| 55 |
## How to Get Started with the Model
|
|
@@ -57,7 +57,7 @@ The model was evaluated using the **MathQA test dataset(2985 examples)** with **
|
|
| 57 |
```python
|
| 58 |
from transformers import AutoTokenizer, AutoModelForCausalLM
|
| 59 |
|
| 60 |
-
local_model_path = "Dasool/
|
| 61 |
tokenizer = AutoTokenizer.from_pretrained(local_model_path)
|
| 62 |
model = AutoModelForCausalLM.from_pretrained(local_model_path)
|
| 63 |
|
|
@@ -101,4 +101,4 @@ The evaluation is based solely on accuracy for a 5-option multiple-choice task.
|
|
| 101 |
|
| 102 |
If you have any questions or feedback, feel free to contact:
|
| 103 |
- Email: dasolcoi@yonsei.ac.kr
|
| 104 |
-
- GitHub: [Repository](https://github.com/Dasol-Choi/
|
|
|
|
| 14 |
- 'LoRA '
|
| 15 |
---
|
| 16 |
|
| 17 |
+
# Model Card for gemmath-2-9b
|
| 18 |
|
| 19 |
This model is based on the Gemma-2-9b architecture and has been fine-tuned using two math problem datasets to improve its accuracy in solving mathematical tasks.
|
| 20 |
|
|
|
|
| 49 |
| Llama-3.2-3b-Instruct | 23.48 |
|
| 50 |
| Qwen2-Math-7B-Instruct| 33.13 |
|
| 51 |
| Qwen2.5-7B-Instruct | 38.69 |
|
| 52 |
+
| **gemmath-2-9b** | **48.91** |
|
| 53 |
|
| 54 |
|
| 55 |
## How to Get Started with the Model
|
|
|
|
| 57 |
```python
|
| 58 |
from transformers import AutoTokenizer, AutoModelForCausalLM
|
| 59 |
|
| 60 |
+
local_model_path = "Dasool/gemmath-2-9b"
|
| 61 |
tokenizer = AutoTokenizer.from_pretrained(local_model_path)
|
| 62 |
model = AutoModelForCausalLM.from_pretrained(local_model_path)
|
| 63 |
|
|
|
|
| 101 |
|
| 102 |
If you have any questions or feedback, feel free to contact:
|
| 103 |
- Email: dasolcoi@yonsei.ac.kr
|
| 104 |
+
- GitHub: [Repository](https://github.com/Dasol-Choi/gemmath-2-9b)
|