This repo contains llamafile version for Phi 2 Electrical Engineering model. This llamafile is created from gguf format of original model's quantized version.
About llamafile
llamafile is the new framework that collapses all the complexity of (large language models) LLMs down to a single-file executable (called a "llamafile") that runs locally on most computers, with no installation. The first release of llamafile is a product of Mozilla’s innovation group and developed by Justine Tunney, about llamafile in short as per introductory post
llamafile lets you turn large language model (LLM) weights into executables.
Say you have a set of LLM weights in the form of a 4GB file (in the commonly-used GGUF format). With llamafile you can transform that 4GB file into a binary that runs on six OSes without needing to be installed.
Basically, llamafile lets anyone distribute and run LLMs with a single file.
Here in github you can find how I created llamafile version from gguf version for other model.
Model Description
From the original model card
This is the adapters from the LoRa fine-tuning of the phi-2 model from Microsoft. It was trained on the STEM-AI-mtl/Electrical-engineering dataset combined with garage-bAInd/Open-Platypus.**
Developed by: STEM.AI
Model type: Q&A and code generation
Language(s) (NLP): English
Finetuned from model [optional]: microsoft/phi-2
Direct Use: Q&A related to electrical engineering, and Kicad software. Creation of Python code in general, and for Kicad's scripting console.
Refer to microsoft/phi-2 model card for recommended prompt format.
GGUF version
The original model was quantized by the Great Mr. Bloke.
As per TheBloke's model card the recommended model phi-2-electrical-engineering.Q5_K_M.gguf quantized with Q5_K_M method this gguf file can take maximum 4.50 GB of RAM.
How to run
To run llamafile version in windows just rename the file by adding .exe at the last and run it as you run any other exe file.
For running in linux run like any binary in linux, just make sure the llamafile is executable by running chmod.
Thats it.
Prompt format
Instruction: <prompt> (without the <>)
Response:
Thank you for using this model.
- Downloads last month
- 0