Loading video...

Video Failed to Load

Go Home

Introducing `gpt-llm-trainer` ✍️ The world's simplest way to train a task-specific LLM. **Just write a sentence describing the model you want.** A chain of AI systems will generate a dataset and train a model for you. And it's open-source:

337,413 views • 2 years ago •via X (Twitter)

11 Comments

Matt Shumer's profile picture
Matt Shumer2 years ago

gpt-llm-trainer is a constrained agent -- meaning its behavior is highly-controlled, leading to better results than open-ended agents. It chains together lots of GPT-4 calls that work together to create a great dataset for you.

Matt Shumer's profile picture
Matt Shumer2 years ago

How it works, in a nutshell: - The user describes the model they want Ex: "A model that writes Python functions" - GPT-4 generates a dataset to train on - We process the dataset, and train a model!

Matt Shumer's profile picture
Matt Shumer2 years ago

It's crazy how quickly you can go from an idea to a fully-trained model with this system. And the resulting models are really good!

Matt Shumer's profile picture
Matt Shumer2 years ago

If you'd like to try it or contribute, check out the Github repo or the Colab notebook:

Teknium (e/λ)'s profile picture
Teknium (e/λ)2 years ago

One suggestion is you should change it from NousResearch/Llama-2 Chat model to llama-2's base model. Training over that terrible chat model doesn't seem to be wise imo

Alex Volkov (Thursd/AI)'s profile picture
Alex Volkov (Thursd/AI)2 years ago

@far__el Matt you're on fire 😂 did you' JUST launch GPT-prompt engineer AND a product hunt launch of personal AI agent?

Matt Shumer's profile picture
Matt Shumer2 years ago

@far__el Haha thanks! I’m only able to do all of this by leveraging AI to help :)

Krishna's profile picture
Krishna2 years ago

This is very cool. What is your recommendation on a use case that is well suited for this (i.e. training a task specific LLM vs. using hosted LLM)?

Matt Shumer's profile picture
Matt Shumer2 years ago

I'm still exploring the possibilities, so right now I'm not too sure! Let me know if you find anything interesting. That said, using a model trained like this would be significantly cheaper than using most hosted LLMs :)

Cooper's profile picture
Cooper2 years ago

Can you give examples of models' performance trained using this tool?

Matt Shumer's profile picture
Matt Shumer2 years ago

I threw this together quickly last night, so I haven't had the chance to test it extensively. But from the few tests I did, if you have GPT-4 generate 200+ examples, it's actually pretty good!

Related Videos

Small Language Models (SML) are the future of AI. "Small" (SML) instead of "Large" (LLM). These small models are highly specialized models with superhuman abilities on specific tasks. Here are two techniques to build these models: • Spectrum • Model Merging I give you a short introduction in the attached video, but here is a quick summary: Spectrum helps us identify the most relevant layers to solve one specific task. We can ignore everything else and focus on fine-tuning these layers. Using Spectrum, we can fine-tune models in a heartbeat. Model Merging combines multiple models into a unique, much better model than any of the individual input models. You can also combine models specialized in different tasks and get a model with multiple abilities. This is the state of the art of productizing models. It's what Arcee.ai's platform does behind the scenes. Arcee collaborated with me on this post and is sponsoring it. There are three main steps to produce a model for your particular use case: 1. You create a dataset by uploading your data. 2. You train a model. At this step, Arcee uses Spectrum and Model Merging to produce a highly specialized model for your task. 3. You can deploy that model to any environment you want. Three important notes: • Training process is 2x faster and 2x cheaper than regular fine-tuning. • Resultant models are smaller and have higher accuracy. • They create these specialized models from open-source models. Check this site so you can fully appreciate how this works: If you want to fine-tune an open-source model, consider Arcee's platform. This is the state of the art.

Santiago

164,162 views • 2 years ago