Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# TinyTextGenerator
|
| 2 |
+
|
| 3 |
+
## Overview
|
| 4 |
+
|
| 5 |
+
TinyTextGenerator is a small causal language model based on the GPT-2 architecture, designed for basic text generation tasks. With only 6 layers, it is lightweight and fast, making it ideal for experimentation, local deployment, or educational use.
|
| 6 |
+
|
| 7 |
+
## Model Architecture
|
| 8 |
+
|
| 9 |
+
- **Model type**: GPT-2 (causal language modeling)
|
| 10 |
+
- **Hidden size**: 768
|
| 11 |
+
- **Number of layers**: 6
|
| 12 |
+
- **Number of attention heads**: 12
|
| 13 |
+
- **Vocabulary size**: 50,257
|
| 14 |
+
- **Context length**: 1024 tokens
|
| 15 |
+
- **Parameters**: ~82M
|
| 16 |
+
|
| 17 |
+
Built using `GPT2LMHeadModel` from the Transformers library.
|
| 18 |
+
|
| 19 |
+
## Usage
|
| 20 |
+
|
| 21 |
+
```python
|
| 22 |
+
from transformers import pipeline
|
| 23 |
+
|
| 24 |
+
generator = pipeline(
|
| 25 |
+
"text-generation",
|
| 26 |
+
model="your-username/TinyTextGenerator"
|
| 27 |
+
)
|
| 28 |
+
|
| 29 |
+
output = generator(
|
| 30 |
+
"The future of AI is",
|
| 31 |
+
max_new_tokens=50,
|
| 32 |
+
do_sample=True,
|
| 33 |
+
top_p=0.95
|
| 34 |
+
)
|
| 35 |
+
|
| 36 |
+
print(output[0]['generated_text'])
|