---
datasets:
- crumb/Wizard-EvolInstruct70k-k4
language:
- en
tags:
- switch_transformers
- llama
- MoE
---
This is the very first testing switchllama model from MoLora2, starting from OpenLlama-3b-v2 and adding 4 experts in the MLP blocks of the model. The experts were trained with QLora and merged properly (in 4bit) after individually training adapters on `gate_proj, up_proj, down_proj`. The 4 expert models were trained on clusters from [crumb/Wizard-EvolInstruct70k-k4](https://huggingface.co/datasets/crumb/Wizard-EvolInstruct70k-k4) then their trained MLP weights were taken and transplanted in a model initialized from OpenLlama-3b with 4 switchtransformer experts. The routers are not trained in this version of the model and are randomly initialized.

Modeling code is not included until this proof-of-concept is entirely trained.