What We Learned from a Year of Building with LLMs (Part III): Strategy
AI/ML
Perplexity Is a Bullshit Machine
A WIRED investigation shows that the AI-powered search startup Forbes has accused of stealing its content is surreptitiously scraping—and making things up out of thin air.
I Will Fucking Piledrive You If You Mention AI Again — Ludicity
Every Way To Get Structured Output From LLMs
A survey of every framework for extracting structured output from LLMs, and how they compare.
Language models on the command-line
I gave a talk about accessing Large Language Models from the command-line last week as part of the Mastering LLMs: A Conference For Developers & Data Scientists six week long …
Rethinking Democracy for the Age of AI - Schneier on Security
ChatGPT is bullshit
The Encyclopedia Project, or How to Know in the Age of AI
In an age when AI regurgitates the blather of meaningless content, seeking its audience in the attention marketplace, it's a small wonder that it is hard to tell what is really real anymore.
Turning the Tables on AI
AI is sold to make us think less... How about using AI not to think less but *more*?
Cost Of Self Hosting Llama-3 8B-Instruct
All blog post from the lytix.ai team
Guitar Pedal Assistant
I believe in human creativity and don’t expect (or at all want) the robots to take over any time soon when it comes to making art in any medium. Every art has technical aspects, though, and i…
Understanding LLM Understanding
Succor borne every minute
Earnest chats with objects are not so unusual. Mark “The Bird” Fidrych, the famed Detroit Tiger, used to stand on the pitching mound whispering to the baseball. Forky, the highly animate utensil from Toy Story 4, once posed deep questions about friendship to a ceramic mug. And many of us have made repeated queries of the Magic 8 Ball despite its limited set of randomly generated answers.
Reader API
Read URLs or search the web, get better grounding for LLMs.
Jina AI Reader
Jina AI provide a number of different AI-related platform products, including an excellent [family of embedding models](https://huggingface.co/collections/jinaai/jina-embeddings-v2-65708e3ec4993b8fb968e744), but one of their most instantly useful is Jina Reader, an API for …
Jina AI Reader
Jina AI provide a number of different AI-related platform products, including an excellent [family of embedding models](https://huggingface.co/collections/jinaai/jina-embeddings-v2-65708e3ec4993b8fb968e744), but one of their most instantly useful is Jina Reader, an API for …
GitHub Copilot Chat: From Prompt Injection to Data Exfiltration · Embrace The Red
Analyzing untrusted code with GitHub Copilot Chat can have malicious side effects and turn Copilot into Copirate!
Gumloop | AI Automation Framework
The no-code platform to build and host AI-powered business automations.
Your brain does not process information and it is not a computer | Aeon Essays
Your brain does not process information, retrieve knowledge or store memories. In short: your brain is not a computer
GitHub - fixie-ai/ultravox
Contribute to fixie-ai/ultravox development by creating an account on GitHub.
Ultravox
Ultravox is "a multimodal Speech LLM built around a pretrained Whisper and Llama 3 backbone". It's effectively an openly licensed version of half of the GPT-4o model [OpenAI demoed](https://openai.com/index/hello-gpt-4o/) (but …
GitHub - ggozad/oterm: a text-based terminal client for Ollama
a text-based terminal client for Ollama.
An Analysis of Chinese LLM Censorship and Bias with Qwen 2 Instruct
A Blog post by lhl on Hugging Face
An Analysis of Chinese LLM Censorship and Bias with Qwen 2 Instruct
Qwen2 is [a new openly licensed LLM](https://qwenlm.github.io/blog/qwen2/) from a team at Alibaba Cloud. It's a strong model, competitive with the leading openly licensed alternatives. It's already ranked 15 on [the …
GenAI Handbook
Dragonfly: A large vision-language model with multi-resolution zoom
Extracting Concepts from GPT-4
A few weeks ago Anthropic [announced they had extracted millions of understandable features](https://simonwillison.net/2024/May/21/scaling-monosemanticity-extracting-interpretable-features-from-c/) from their Claude 3 Sonnet model. Today OpenAI are announcing a similar result against GPT-4: We …
Supervision and truth
A diagram I found online about distinguishing facts and value judgments. Seems easy enough! For more than a decade the dominant training methodology in...
What the AI boom is getting wrong (and right), according to Hugging Face’s policy chief
Irene Solaiman on tackling racial bias, language equity, consent, and why there are so many AI girlfriends.
AI Integration and Modularization
Breaking down the Big Tech AI landscape through the lens of integration and modularization