Phi-3-medium-4k-instruct-ko-poc-gguf-v0.1

Model Details

This model converted the daekeun-ml/Phi-3-medium-4k-instruct-ko-poc-v0.1 to gguf 4-bit format.

For detailed instructions, please refer to Microsoft's official repo.

Dataset

The dataset used for training is as follows. To prevent catastrophic forgetting, I included non-Korean corpus as training data. Note that we did not use all of the data, but only sampled some of it. Korean textbooks were converted to Q&A format. The Guanaco dataset has been reformatted to fit the multiturn format like <|user|>\n{Q1}<|end|>\n<|assistant|>\n{A1}<|end|>\n<|user|>\n{Q2}<|end|>\n<|assistant|>\n{A2}<|end|>.

• Korean textbooks (https://huggingface.co/datasets/nampdn-ai/tiny-codes)
• Korean translation of Guanaco (https://huggingface.co/datasets/nlpai-lab/openassistant-guanaco-ko)
• Guanaco Sharegpt style (https://huggingface.co/datasets/philschmid/guanaco-sharegpt-style)

How to Get Started with the Model using Ollama

1. Install Ollama:

curl -fsSL https://ollama.com/install.sh | sh

2. Get the Modelfile:

huggingface-cli download daekeun-ml/Phi-3-medium-4k-instruct-ko-poc-gguf-v0.1 Modelfile_q4 --local-dir /path/to/your/local/dir

3. Build the Ollama Model: Use the Ollama CLI to create your model with the following command:

ollama create phi3-ko -f Modelfile_q4

4. Run the model:

ollama run phi3-ko What is Machine Learning?

Notes

License

apache 2.0; The license of phi-3 is MIT, but I considered the licensing of the dataset and library used for training.

Caution

This model was created as a personal experiment, unrelated to the organization I work for. The model may not operate correctly because separate verification was not performed. Please be careful unless it is for personal experimentation or PoC (Proof of Concept)!