Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -56,3 +56,20 @@ The following hyperparameters were used during training:
|
|
| 56 |
- Pytorch 2.4.0+cu121
|
| 57 |
- Datasets 2.16.1
|
| 58 |
- Tokenizers 0.19.1
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 56 |
- Pytorch 2.4.0+cu121
|
| 57 |
- Datasets 2.16.1
|
| 58 |
- Tokenizers 0.19.1
|
| 59 |
+
Vanilla model performance on frozenlake prompts dataset:
|
| 60 |
+
- Exact match: 0.00%
|
| 61 |
+
- F1: 17.75%
|
| 62 |
+
|
| 63 |
+
Fine-tuned model performance on frozenlake prompts dataset:
|
| 64 |
+
- Exact match: 21.00%
|
| 65 |
+
- F1: 63.13%
|
| 66 |
+
|
| 67 |
+
## Model performance on frozenlake prompts dataset
|
| 68 |
+
|
| 69 |
+
Vanilla model performance:
|
| 70 |
+
- Exact match: 0.00%
|
| 71 |
+
- F1: 17.75%
|
| 72 |
+
|
| 73 |
+
Fine-tuned model performance:
|
| 74 |
+
- Exact match: 21.00%
|
| 75 |
+
- F1: 63.13%
|