Phi-Hybrid-1.5B is an experimental hybrid language model that merges the capabilities of Phi-1 and Phi-1.5 from Microsoft. Built using MergeKit, this model is designed to enhance performance while maintaining efficiency, making it a powerful tool for text generation.

🔗 Created by: [Matteo Khan ]
🎓 Affiliation: Apprentice at TW3 Partners (Generative AI Research) 📍 License: MIT

🔗 Connect with me on LinkedIn
🔍 Model on Hugging Face

🧠 Model Details

Model Type: Hybrid Language Model (Merged)
Parent Models:
- Phi-1
- Phi-1.5
Merging Technique: Linear Merge (MergeKit)

🎯 Intended Use

This model is intended for research and experimentation in hybrid model optimization. Potential use cases include:

✅ Text Generation
✅ Conversational AI
✅ Code Assistance
✅ Creative Writing
✅ Exploration of Model Merging Effects

⚠️ Limitations & Considerations

While Phi-Hybrid-1.5B enhances certain capabilities, it also inherits limitations from its parent models:

❌ May generate inaccurate or misleading information
⚠️ Potential for biased, offensive, or harmful content
🔄 Merging may introduce unpredictable behaviors
📉 Performance may vary across different tasks

🔬 Merging Process & Configuration

This is not a newly trained model, but rather a merge of existing models using the following configuration:

yaml

merge_method: linear
dtype: float16
models:
  - model: "microsoft/phi-1"
    parameters:
      t: 1.0
      weight: 0.6
  - model: "microsoft/phi-1_5"
    parameters:
      t: 1.0
      weight: 0.4
parameters:
  normalize: true
  int8_mask: false
layers:
  - pattern: "model.*"

📊 No formal evaluation has been conducted yet. Users are encouraged to benchmark and share feedback!

🌍 Environmental Impact

By utilizing model merging rather than training from scratch, Phi-Hybrid-1.5B significantly reduces computational and environmental costs.

🚀 How to Use

python

from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "YourProfile/Phi-Hybrid-1.5B"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)

# Example usage
prompt = "Explain the theory of relativity in simple terms."
inputs = tokenizer(prompt, return_tensors="pt")
outputs = model.generate(**inputs, max_length=200)
response = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(response)

📜 Citation & References

If you use Phi-Hybrid-1.5B in your research, please cite the parent models:

📝 Phi-1

bibtex

@misc{phione,
      title={Phi-1: A Small-Scale Language Model for Reasoning},
      author={Microsoft Research},
      year={2023},
      url={https://huggingface.co/microsoft/phi-1}
}

📝 Phi-1.5

📩 Feedback & Contact: Reach out via Hugging Face.

🎉 Happy Experimenting! 🚀

MatteoKhan/phi-1-1.5-merged

🚀 Phi-Hybrid-1.5B: Merging Efficiency & Power

📌 Overview