rjb7731

Isn't this similar to the Deepmind paper on long form factuality posted a few days ago?

https://arxiv.org/abs/2403.18802

https://github.com/google-deepmind/long-form-factuality/tree...

I've previously noticed when playing with GPT-4 it can sometimes 'autocomplete' on different sections of the text its feeding back, sometimes what looks like 4 or more different sections. Might be unrelated but is this MoE in action or them streaming the response in some way?

Its interesting this paper & the Orca LLM paper from Microsoft are using GPT3/4 model outputs to train 'powerful' models. Big question is will they allow your average joes/businesses to do the same on their own data? Doubt it, considering it breaks OAIs terms of use at present. Will opensource lead the way on this? Bring on Llama-v2

Thanks Jeremy for putting this course out for everyone, looking forward to completing it.

Seems not too bad, you can tell its been trained on shorter responses.

Example #1

Input: explain a neural network in simple terms

Alpaca-Output: A neural network is a type of machine learning algorithm that is made up of layers of interconnected nodes (aka neurons). It is inspired by the way the brain works and uses interconnected nodes to learn patterns in data and make predictions. Neural networks can be used to perform various tasks such as classification, regression, and speech/image recognition.

Example #2

Input: explain how transformers work in artificial intelligence

Alpaca-Output: Transformers are a type of artificial intelligence algorithm used in deep learning. They are responsible for taking input data and converting it into a different data representation, called the "output space". Transformers do this by using a series of mathematical operations, allowing them to map input data onto a new, desired output. This allows the artificial intelligence system to perform tasks such as natural language processing and image recognition. Transformers are able to learn from data and can be used to build more efficient models for a variety of tasks

Example #3

Input: how do i make a request to the chatgpt api with python?

Alpaca-Output: The API can be accessed using the Python language. You can use requests library to make API requests, for example:

import requests

# URL of the chatgpt API url = 'https://api.chatgpt.com/v1/messages'

# Post a new message data = { 'text': 'Hello World' }

# Send the request and get the response response = requests.post(url, data=data) print(response.text)

With the 7B none-instruction tuned you have to really pre-prompt it to get good answers to things so like:

SYSTEM: You are an AI research assistant. You use a tone that is technical and scientific.

USER: Hello, who are you?

ASSISTANT: Hi! I am an AI research assistant. How can I help you today?

USER: Can you tell me something about owls.

The inference on the gradio demo seems pretty slow, about 250 seconds for a request. Maybe I am too used to the 4-bit quant version now ha!

if you look in the dev tools you will see a request to a 'join' file when you click the agree button it adds you to a queue. You can watch where you are up to in the dev tools.

Even with patience, I can see nothing happening in the dev tools when i click agree?

Edit: From looking when you click agree you can see a request to a 'join' file with a 101 status, looks like a queuing system and I'm 219th..

interesting, looks like the web demo doesn't work at the moment though. The prompt.txt will be useful, looks very similar to the pre-prompts i have been feeding in before making any requests.

Looks like its no bueno on google colab, also it takes prompts in via input() too rather then a command line argument.

I wish him the best, I find his streams incredibly useful and they motivate me to be better at solving problems and coding. Hopefully he still does work on AGI and makes tinygrad even cooler.