Highlights
- Pro
Pinned Loading
-
creating_Phi2_MoE_using_mergekit.md
creating_Phi2_MoE_using_mergekit.md 1# microsoft/phi-2 for creating Mixture of Experts (MoE)2The [microsoft/phi-2](https://huggingface.co/microsoft/phi-2) is a small language model with 2.7billion parameters. Because of its small size, opensource license and thanks to finetuning techqniques like QLoRA, one can (fairly) quickly finetune a base model for performing downstream tasks and creating an expert phi-2 model. It would be interesting to combine the individual experts into a Mixture of Experts (MoE) to make the MoE perform the tasks of the individual experts. Follow the steps below to create your own version of a MoE based out of phi-2.
3- Special mention to [Maxime Labonne](), [Aratako](https://github.com/Aratako), [Paul Ilioaica](https://github.com/paulilioaica) for showing the opensource community that the mergekit can be tweaked to make a MoE out of phi-2 experts.
4- Big shoutout to [Charles O. Goddard](https://github.com/cg123), the author of mergekit for creating and letting us play with [mergekit](https://github.com/arcee-ai/mergekit)
5 -
berts.cpp-on-android
berts.cpp-on-android PublicForked from yilong2001/berts.cpp
CPP inference of Bert-model family based on GGML, supporting sentence classification models, seq2seq text generation models, etc.
C 1
-
SORT-java
SORT-java PublicA Java implementation of the SORT (Simple Online and Realtime Tracking) algorithm, ready to integrate into Android applications.
Java
-
mergekit
mergekit PublicForked from arcee-ai/mergekit
Tools for merging pretrained large language models.
-
FREEkMapper
FREEkMapper PublicA powerful projection mapping application built with Python, Tkinter, and OpenGL. Designed for live performance and installation art, Freakmapper allows you to map video and images onto physical su…
Python 1
If the problem persists, check the GitHub status page or contact support.




