Update README.md
Browse files
README.md
CHANGED
@@ -10,37 +10,34 @@ tags:
|
|
10 |
# ArabianGPT Model Overview
|
11 |
|
12 |
## Introduction
|
13 |
-
ArabianGPT
|
|
|
|
|
14 |
|
15 |
## Key Features
|
16 |
-
|
17 |
-
|
18 |
-
|
19 |
-
|
20 |
-
|
21 |
-
| **Layers** | 12 |
|
22 |
-
| **Model Attention Layers**| 12 (MAL) |
|
23 |
-
| **Context Window Size** | 768 tokens |
|
24 |
|
25 |
## Training
|
26 |
-
|
27 |
-
|
28 |
-
|
29 |
-
|
30 |
-
|
31 |
-
|
32 |
-
|
33 |
-
|
34 |
-
|
35 |
-
| **Training Duration**| 3 days |
|
36 |
-
| **Performance** | Final loss of 3.97 |
|
37 |
|
38 |
|
39 |
## Role in ArabianLLM Initiatives
|
40 |
-
ArabianGPT
|
41 |
|
42 |
## Usage
|
43 |
-
Suitable for Arabic text generation tasks. Example usage with Transformers
|
44 |
```python
|
45 |
from transformers import pipeline
|
46 |
|
|
|
10 |
# ArabianGPT Model Overview
|
11 |
|
12 |
## Introduction
|
13 |
+
ArabianGPT-0.1B, developed under the ArabianLLM initiatives, is a specialized GPT-2 model optimized for Arabic language modeling.
|
14 |
+
It's a product of the collaborative efforts at Prince Sultan University's Robotics and Internet of Things Lab, focusing on enhancing natural language modeling and generation in Arabic.
|
15 |
+
This model represents a significant stride in LLM research, specifically addressing the linguistic complexities and nuances of the Arabic language.
|
16 |
|
17 |
## Key Features
|
18 |
+
- **Architecture**: GPT-2
|
19 |
+
- **Model Size**: 134 million parameters
|
20 |
+
- **Layers**: 12
|
21 |
+
- **Model Attention Layers (MAL)**: 12
|
22 |
+
- **Context Window Size**: 768 tokens
|
|
|
|
|
|
|
23 |
|
24 |
## Training
|
25 |
+
- **Dataset**: Abu Elkhiar Corpus
|
26 |
+
- **Data Size**: 15.5 GB
|
27 |
+
- **Words**: 237.8 million
|
28 |
+
- **Tokenizer**: Aranizer 64K
|
29 |
+
- **Tokens**: Over 1.75 billion
|
30 |
+
- **Hardware**: 2 NDIVIA A100 GPUs
|
31 |
+
- **Training Scale**: 7.5 million examples
|
32 |
+
- **Training Duration**: 3 days
|
33 |
+
- **Performance**: Final loss of 3.97
|
|
|
|
|
34 |
|
35 |
|
36 |
## Role in ArabianLLM Initiatives
|
37 |
+
ArabianGPT-0.1B (Base Model) is crucial for advancing Arabic language processing, addressing challenges unique to Arabic morphology and dialects.
|
38 |
|
39 |
## Usage
|
40 |
+
Suitable for Arabic text generation tasks. Example usage with Transformers Pipeline:
|
41 |
```python
|
42 |
from transformers import pipeline
|
43 |
|