Test Model

This is a model created for testing purposes. It is one of the smallest models available on Hugging Face, designed to experiment with the capabilities of small/tiny models.

Model Structure

The model is structured to handle text generation tasks. It is trained on a small dataset that consists of user prompts and AI responses, allowing it to generate text based on given inputs.

Usage

You can use this model for text generation tasks, particularly for testing and experimentation with small models. It is not intended for production use or serious applications.

Prompt format

<|user|>Your input prompt here<|ai|>The model's response will be generated here<|stop|>

Example Usage

Transformers

from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "Malte0621/test-model"
model = AutoModelForCausalLM.from_pretrained(model_name, gguf_file="model.gguf")
tokenizer = AutoTokenizer.from_pretrained(model_name)
prompt = "<|user|>Hi there!<|ai|>"
response = model.generate(tokenizer(prompt, return_tensors="pt").input_ids)
print(tokenizer.decode(response[0], skip_special_tokens=True))

llama.cpp

./llama-cli -m model.gguf -p "<|user|>Hi there!<|ai|>" -n 128 -no-cnv --rope-freq-scale 0.125

LM Studio

https://model.lmstudio.ai/download/Malte0621/test (Make sure to set the RoPE frequency scale to 0.125 in the model settings.)

License

This model is released under the Fair Noncommercial Research License.

Citation

If you use this model in your research, please cite it as follows:

@misc{test-model,
  author = {Malte0621},
  title = {Test-Model},
  year = {2025},
  url = {https://huggingface.co/Malte0621/test}
}

Acknowledgements

This model was created as part of a personal project to explore the capabilities of small language models. It is not affiliated with any organization or commercial entity.