Practical analysis of ChatGPT, RAG, fine-tuning, tokenisation and AI products.
2026-09-06
GPT-6 Astra deserves attention for its economics and for how difficult some of its visible reasoning is to follow. Some of its visible thinking I observed looks like dense word-soup with little spacing and barely legible partial sentences. It raises a practical question for anyone building AI products. If a model does better work while showing less comprehensible reasoning, how do we verify it?
2024-04-12
128k tokens are 96k words in English for ChatGPT 3.5 and 4. The ratio is estimated to be 0.75 words per token. However, the answer is not straightforward for other languages, but we can approximate it using English. Confused? I was. Let me explain.
2023-11-29
Large-language models (LLMs) are great generalists, but modifications are required for optimisation or specialist tasks. The easiest choice is Retrieval Augmented Generation (RAG) with in-context, which has become easier with OpenAI's custom GPTs. The hardest and most unattainable option is creating a specialist LLM from scratch. But you can achieve much of it by simply adjusting an existing LLM like ChatGPT, a process called fine-tuning. You can do that with open-source or commercial models...
2023-11-23
Recently, OpenAI released GPT4 turbo preview with 128k at its DevDay. That addresses a serious limitation for Retrieval Augmented Generation (RAG) applications, which I described in detail for Llamar.ai. That amounts to nearly 200 pages of text, assuming approximately 500 words per page and 0.75 words per token and 2¹⁷ tokens.
2023-11-09
Today, I received access to the new custom GPT feature on ChatGPT, and it appears to do what Sam Altman demonstrated. The implications are far-reaching, beyond the death of the RAG business model. How can OpenAI achieve and capitalise on Omniscience in a world of organisational data silos and increasingly defensive content creators and legislators?
2023-11-07
OpenAI's DevDay announcement yesterday addresses issues I wrote about in the infeasibility of RAG after building Llamar.ai this summer. Did I get it wrong? Working through the details will take some time, but interesting immediate observations are apparent.
2023-09-27
Over four months, I created a working retrieval-augmented generation (RAG) product prototype for a sizeable potential customer using a Large-Language Model (LLM). It became a ChatGPT-like expert agent with deep, up-to-date domain knowledge and conversational skills. But I am shutting it down instead of rolling it out. Here is why, how I got there and what it means for the future.
2023-04-12
Learn to harness the potential of ChatGPT4, your virtual programming partner, with nine prompting tips. Improve your programming skills by communicating clearly, engaging in conversation, using the proper syntax, and iterating on complexity. Keep context fresh, and ChatGPT4 will be invaluable in your coding journey.
2023-04-05
Test-driven development in JavaScript with ChatGPT-4 works. An example demonstrates it using a precise description and refined prompt engineering.
2023-04-04
Open Source libraries offer user documentation. But expert users and contributors have a deeper understanding of the inner workings stemming from a mental model and architecture derived from deep dives into the code. That understanding and model are helpful to employ the library more effectively, debug issues when using it, and teach interesting concepts on how to structure complex reusable code.
2023-02-11
Microsoft could follow Google's $100bn loss. I tried the new Bing Chat (ChatGPT) feature, which was great until it went disastrously wrong. It even started arguing with me while being wrong and making source code up.
2023-02-07
The Battle of the AI Chatbots Begins: Google's Bard Takes on ChatGPT.
2023-02-07
ChatGPT is a state-of-the-art language model developed by OpenAI, utilising the Transformer model and fine-tuned through reinforcement learning to produce accurate and ethical text responses.
2023-02-03
ChatGPT can combine Data with natural language and has extensive information about most subjects. That lends itself to novel applications like creating informative data dictionaries.
2023-02-02
Programming with ChatGPT using an iterative approach is difficult, as I have demonstrated previously. Maybe ChatGPT can benefit from Test-driven development (TDD). Could it aid LLMs as it does humans?
2023-01-29
Can ChatGPT help you develop software in Python? Let us ask ChatGPT to write code to query AWS Athena to test if and how we can do it step-by-step.
2023-01-25
ChatGPT and similar language models have recently been gaining attention for their potential to revolutionise code generation and enhance developer productivity. I was curious to see what all the hype was about, so I decided to try it out for some development work.
2022-12-04
OpenAI's ChatGPT has made the news recently as a next-generation conversational agent. It has a surprising breadth which made me wonder, could OpenAI generate specific technology content good enough to post, and what would that imply for the future?
2022-06-20
Insurance works because it shares costs in the face of uncertainty. What happens when Tesla removes uncertainty and distributes cost seemingly more fairly? First partially and eventually wholly? Will insurance fail, doing more harm than good?