About Orca-2 The fine folk at Microsoft Research have recently published Orca 2, a new small large language model and apparently, it’s quite good! Just look at the test results below – on average, both the 7B and the 13B variants are significantly better than Llama-2-Chat-70B, with Orca-2-13B superseding even WizardLM-70B. Pretty cool! 🚀 I also love the idea behind it: prompting a big large language model (in our case GPT-4) to answer some rather convoluted logic questions while aided by some very specific system prompts, and then fine-tune a smaller model (Llama-2-7B and 13B respectively) on just the question and answer pairs, leaving out the detailed system prompts. ...

These are the slides and notebook I’ve used during my talk on how to build an Internet-connected search assistant almost from scratch. AKA Poor Man’s BingChat. First time I talked about it was at Codecamp Iasi, where it’s gotten a lot of positive feedback, plus it was awesome to share the stage with established speakers (and personal heroes of mine) like Mark Richards, Venkat Subramaniam, Eoin Woods, and Dylan Beattie. Yes, you can see them in the hero picture 😱. ...

Recently, I was curious to see how easy it would be to run run Llama2 on my MacBook Pro M2, given the impressive amount of memory it makes available to both CPU and GPU. This led me to the excellent llama.cpp, a project focused on running simplified versions of the Llama models on both CPU and GPU. The process felt quite straightforward except for some instability in the llama.cpp repo just as I decided to try it out, and which has been fixed in the meantime. Incidentally, this prompted me to document the whole process, just in case I want to do it again in the future. ...

An experiment with prompt injecting Bing Chat – successfully changing its persona, exploring data extraction potential, limitations, and future implications.

My top 3 tips for working better, faster, and just a bit stronger with Azure ML Compute Instances

A quickstart guide to deploying machine learning models in production using Azure Machine Learning’s managed online endpoints

A guide to creating GPU compute instances on Azure ML, installing Stable Diffusion, and running AUTOMATIC1111’s Web UI.

Because life’s too short to deploy things manually

A glimpse of the upcoming paradigm shift in how we do development

The issue with machine learning pipelines is that they need to pass state from one step to another. When this works, it’s a beautiful thing to behold. When it doesn’t, well, it’s not pretty, and I think the clip below sums this up pretty well. made a Rube Goldberg machine pic.twitter.com/gWRNnmm5Ic — COLiN BURGESS (@Colinoscopy) April 30, 2020 Azure ML Pipelines are no stranger to this need for passing data between steps, so you have a variety of options at your disposal. This means it’s not always easy to find the best one, and I’ve often seen people confused when trying to pick the best option. So I wrote this article to try and clear some of that confusion. ...