83 lines
1.8 KiB
Markdown
83 lines
1.8 KiB
Markdown
---
|
|
license: apache-2.0
|
|
datasets:
|
|
- rahul77/rahul-gpt2-1k
|
|
language:
|
|
- en
|
|
base_model: openai-community/gpt2
|
|
pipeline_tag: text-generation
|
|
library_name: transformers
|
|
tags:
|
|
- text-generation
|
|
- GPT-2
|
|
- fine-tuned
|
|
- language-model
|
|
- transformers
|
|
---
|
|
|
|
# GPT-2 Fine-Tuned Model
|
|
|
|
This is a fine-tuned version of the GPT-2 model designed for text generation tasks. The model has been fine-tuned to improve its performance on generating coherent and contextually relevant text.
|
|
|
|
## Model Details
|
|
|
|
- **Model Name:** GPT-2 Fine-Tuned
|
|
- **Base Model:** gpt2
|
|
- **Architecture:** GPT2LMHeadModel
|
|
- **Tokenization:** Supported
|
|
- `pad_token_id`: 50256
|
|
- `bos_token_id`: 50256
|
|
- `eos_token_id`: 50256
|
|
|
|
## Supported Tasks
|
|
|
|
This model supports the following task:
|
|
|
|
- **Text Generation**
|
|
|
|
## Configuration
|
|
|
|
### Model Configuration (config.json)
|
|
|
|
- **Hidden Size:** 768
|
|
- **Number of Layers:** 12
|
|
- **Number of Attention Heads:** 12
|
|
- **Vocab Size:** 50257
|
|
- **Token Type IDs:** Not used
|
|
|
|
### Generation Configuration (generation_config.json)
|
|
|
|
- **Sampling Temperature:** 0.7
|
|
- **Top-p (nucleus sampling):** 0.9
|
|
- **Pad Token ID:** 50256
|
|
- **Bos Token ID:** 50256
|
|
- **Eos Token ID:** 50256
|
|
|
|
## Usage
|
|
|
|
To use this model for text generation via the Hugging Face API, use the following Python code snippet:
|
|
|
|
```python
|
|
import requests
|
|
|
|
api_url = "https://api-inference.huggingface.co/models/rahul77/gpt-2-finetune"
|
|
headers = {
|
|
"Authorization": "Bearer YOUR_API_TOKEN", # Replace with your Hugging Face API token
|
|
"Content-Type": "application/json"
|
|
}
|
|
|
|
data = {
|
|
"inputs": "What is a large language model?",
|
|
"parameters": {
|
|
"max_length": 50
|
|
}
|
|
}
|
|
|
|
response = requests.post(api_url, headers=headers, json=data)
|
|
|
|
if response.status_code == 200:
|
|
print(response.json())
|
|
else:
|
|
print(f"Error: {response.status_code}")
|
|
print(response.json())
|