This tutorial series will show you how to build an end-to-end data flywheel for Large Language Models (LLMs).

We will be summarising arXiv abstracts.

How to:

• Build a training set with GPT-4 or GPT-3.5
• Fine-tune an open-source LLM
• Create a set of Evals to evaluate the model.
• Collect human feedback to improve the model.
• Deploy the model to an inference endpoint.

• wandb for experiment tracking. This is where we will record all our artifacts (datasets, models, code) and metrics.
• modal for running jobs on the cloud.
• huggingface for all-things-LLM.
• argilla for labelling our data.

In this tutorial, we will use GPT-3.5 to generate a training set for summarisation task.

modal run src/llm_stack/scripts/build_dataset_summaries.py

Found any mistakes or want to contribute? Feel free to open a PR or an issue.