Skip to content

upadraj/In-context

Repository files navigation

dl-project-llm

Introduction

We run Efficient LLM Context Distillation model below:

how to install:

 conda env create -f relu_ranger.yaml
 conda activate relu_ranger

If in Visual Studio, you can CTRL + P => select python interpretert to select relu_ranger as your default environment.

Run jupyter notebook and open up the relevant file: opt-125m.ipynb or teacher_student.ipynb

Run all code to get model outputs.

Interpretation:

The project implements context distillation by training a student model on a KL-divergence loss derived from a teacher model. Additionally, LoRA (Low-Rank Adaptation) is incorporated.

Models

Running the Model in Google Colab

Upload run_models.ipynb notebook and its dependencies (contained in context_utils.py, training_utils.py, and data_utils.py) on Google Colab.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published