Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,36 @@
|
|
| 1 |
-
---
|
| 2 |
-
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language: en
|
| 3 |
+
license: mit
|
| 4 |
+
library_name: transformers
|
| 5 |
+
tags:
|
| 6 |
+
- gpt2
|
| 7 |
+
- legal
|
| 8 |
+
- text-generation
|
| 9 |
+
- nepal
|
| 10 |
+
- transformer
|
| 11 |
+
- pytorch
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
# Nepal Legal GPT-2 (From Scratch Implementation)
|
| 15 |
+
|
| 16 |
+
This is a custom GPT-2 style transformer model trained from scratch on Nepal's legal documents. The model was implemented entirely in PyTorch without relying on pre-trained weights, specifically designed to understand and generate text related to Nepal's legal domain.
|
| 17 |
+
|
| 18 |
+
## Model Details
|
| 19 |
+
|
| 20 |
+
- **Model Architecture**: GPT-2 style Transformer
|
| 21 |
+
- **Parameters**: ~1 million
|
| 22 |
+
- **Context Length**: 128 tokens
|
| 23 |
+
- **Layers**: 6
|
| 24 |
+
- **Attention Heads**: 6
|
| 25 |
+
- **Embedding Dimension**: 384
|
| 26 |
+
- **Vocabulary Size**: 50,257 (GPT-2 tokenizer)
|
| 27 |
+
- **Training Data**: Nepal Legal QA English Dataset
|
| 28 |
+
- **License**: MIT
|
| 29 |
+
|
| 30 |
+
## Training Data Experimental
|
| 31 |
+
|
| 32 |
+
The model was trained on a specialized dataset of Nepal's legal documents in English, containing:
|
| 33 |
+
- Legal questions and answers
|
| 34 |
+
- Procedural instructions
|
| 35 |
+
- Legal definitions and explanations
|
| 36 |
+
- Court procedures and regulations
|