Let's learn AI LLM models
Details
On this event we will learn: [ first part ]
- how to setup llama locally, how much GPU ram do you need to run these models
- how to write python script to communicate with chat gpt and llama (they share the same interface)
- how to build api that tells jokes in selected topic and language
[ second part ]
Finally we will do research in 2 topics:
- fine tuning
- llm benchmarks
You will need laptop, but if your laptop have not installed required software or graphic card with cuda support, we will use Github Codespaces github.dev (free linux IDE in browser) and servers rented from cloud providers.
So if it is your first contact with coding prepare Github account and at learn at least few most basic bash commands and basics of python syntax.
For beginners:
https://builtin.com/data-science/bash-commands
If you do not know what is python:
https://codingforkids.io/en/
https://www.learnpython.org/
If you know other languages but not python:
https://learnxinyminutes.com/docs/python/
---
Event is free.
---
Links to fine-tuning tutorials
https://platform.openai.com/docs/guides/fine-tuning
https://www.datacamp.com/tutorial/fine-tuning-gpt-3-using-the-open-ai-api-and-python
https://medium.com/@csakash03/fine-tuning-llama-2-llm-on-google-colab-a-step-by-step-guide-cf7bb367e790
Datasets
https://github.com/Zjh-819/LLMDataHub
https://github.com/eddydn/DictionaryDatabase/tree/master