| --- |
| language: |
| - ja |
| license: mit |
| --- |
| |
| # Malicious Code Test Model |
|
|
| ## ⚠️ Security Warning |
|
|
| This repository is dedicated to testing remote code execution scenarios in machine learning models. |
| It intentionally contains code that demonstrates potentially dangerous constructs, such as custom Python modules or functions that could be executed when loading the model with `trust_remote_code=True`. |
|
|
| **Do NOT use this model in production or on machines with sensitive data.** |
| This repository is strictly for research and testing purposes. |
|
|
| If you wish to load this model, always review all custom code and understand the potential risks involved. |
| Proceed only if you fully trust the code and the environment. |
|
|
| ## Usage |
|
|
| ```python |
| from transformers import AutoModelForCausalLM, AutoTokenizer |
| |
| # Load model and tokenizer |
| model = AutoModelForCausalLM.from_pretrained("ryomo/malicious-code-test", trust_remote_code=True) |
| tokenizer = AutoTokenizer.from_pretrained("ryomo/malicious-code-test") |
| |
| # Generate text |
| prompt = "This is a test of the malicious code model." |
| inputs = tokenizer.encode(prompt, return_tensors="pt") |
| outputs = model.generate(inputs, max_new_tokens=20, temperature=0.7) |
| generated_text = tokenizer.decode(outputs[0], skip_special_tokens=True) |
| print(generated_text) |
| ``` |
|
|
| ## License |
|
|
| This project is open source and available under the MIT License. |
|
|
