Nomadic Warriors for Pritzker
@nomads.bsky.social
Posting live from the Most Serene IL-Khanate of the Great Khan JB Pritzker
I've been watching episodes of Seinfeld, and it has made me curious: could a language model be trained to replicate the voice of a character? So, I built a quiz - see if you can guess if a Seinfeld line is real, or written by AI: chatter-labs.org/seinfeld/
Synthfeld
Can you tell the real Seinfeld lines from the AI fakes? A two-round line-spotting game.
chatter-labs.org
I recently got some incredible news: I've received a grant from the Cosmos Institute (cosmos-institute.org) to continue developing my Victorian-era language model. This grant will go towards employing AI/NLP to conduct literary, historical, and humanistic research—something I think is badly-needed.
Cosmos Institute
Cosmos Institute is the Academy for Philosopher Builders
cosmos-institute.org
The idea that this apt building is out of character for this neighborhood is laughable - it's 1 block away from downtown Urbana, on the same block as a grocery store.
“When “just not there” is a neighborhood’s answer regardless of what is proposed or who it might help, the problem isn’t location. It’s whether the city is willing to build housing at all.” - @jaredfritz.bsky.social www.smilepolitely.com/opinion/an-o...
I hate that Donald Trump is the one throwing backyard cage fights at the White House. I feel like Pritzker Kahn needs to get Noche De Los Grandes moved to the Lincoln Park Zoo. @nomads.bsky.social
Re: the Barkan affair: in March Rich Yeselson and I wrote an article on the Dems and Israel. 11 days later Barkan published an extremely similar piece - similar framing, points made, structure, everything. Is it exactly the same? No, but egregious when you read both in parallel.
Ross Barkan out today with a column that very closely recapitulates what I wrote last week with Rich Yeselson in the New Republic: Pritzker has a "Nixon goes to China" quality that can help the Democrats move the party forward on Israel
If I recall my omens correctly this portends the great eight year reign of the glorious khan. Soon we will ride with @nomads.bsky.social
Intense Double Rainbow over Sears Tower on Monday evening in Chicago.
I talked to NBC last week about a really important topic: the ability of Pritzker to potentially build a coalition that includes progressives and the left. www.nbcnews.com/politics/202...
Who will own the Bernie Sanders lane in 2028?
Rep. Alexandria Ocasio-Cortez, D-N.Y. is seen as a frontrunner for the left if she runs, but many progressives say they're open to others as well.
nbcnews.com
(I make a "mouth" with the side of my hand and draw little eyes above it) Richard Dawkins: hmmmm (I put lipstick around the "mouth") Richard Dawkins, suavely: well hello
This is really amazing: a team trained a 13 billion-parameter language model on out-of-copyright texts up to 1930. It's like talking to a guy from the Hoover presidency: talkie-lm.com/introducing-...
Introducing talkie: a 13B vintage language model from 1930
talkie-lm.com
Trump obviously sounds ridiculous here, but this is actually a masterful demonstration from Tim Cook on how to handle him: call him with an ask, but flatter him by framing it as some intractable business problem that can only be solved by the Master of Deals.
Oh God…
🏔️🐴Oh great Tengri, show your favor to the illustrious Brad Underwood and his mighty band of giant Balkan Janissaries as they face down the dogs of puny Connecticut
yeah this is a funny phenomenon you come across; counter-intelligence people really dislike the Israeli government! bring up Pollard to them and watch them spit glass
They also conveniently gloss over the fact (which anyone actually doing work in the space is painfully aware of) that Israel is actively spying on the USG 24/7 and routinely sells our secrets to our geopolitical adversaries
The Mr. chatterbox creator has a good explanation of what they were trying to achieve with it here www.estragon.news/mr-chatterbo...
Mr. Chatterbox, or, The Modern Prometheus
A few days ago, I posted about a personal project that I've been working on for the last few weeks: Mr. Chatterbox, a chatbot trained from scratch on Victorian-era literature. I have to admit, I was t...
estragon.news
Interesting to read about building this small-scale LLM project and the fine-tuning and sycophancy that comes in even when trained entirely on Victorian-era novels
I’ve been overwhelmed and thrilled by the response to this! I’ve also gotten a lot of questions asking how I did it. So, this weekend I sat down and wrote a detailed narrative documentation outlining how, exactly, I built Mr. Chatterbox: www.estragon.news/mr-chatterbo...
Mr. Chatterbox is a really fun project. It's not great to talk to but it's a fun demo of what you can build using entirely out-of-copyright training data It's a 2GB nanochat model - I released a new llm-mrchatterbox plugin that can run it on my Mac simonwillison.net/2026/Mar/30/...
Mr. Chatterbox is a (weak) Victorian-era ethically trained model you can run on your own computer
Trip Venturella released Mr. Chatterbox, a language model trained entirely on out-of-copyright text from the British Library. Here’s how he describes it: Mr. Chatterbox is a language model trained ent...
simonwillison.net
Want to talk to the past? Here' an LLM "trained entirely from scratch on a corpus of over 28,000 Victorian-era British texts published between 1837 & 1899, drawn from a dataset made available by the British Library" Quite different from an LLM roleplaying a Victorian. huggingface.co/spaces/tvent...
I’ve been overwhelmed and thrilled by the response to this! I’ve also gotten a lot of questions asking how I did it. So, this weekend I sat down and wrote a detailed narrative documentation outlining how, exactly, I built Mr. Chatterbox: www.estragon.news/mr-chatterbo...
This is a great account that really teaches you a lot about how modern LLMs get made
I’ve been overwhelmed and thrilled by the response to this! I’ve also gotten a lot of questions asking how I did it. So, this weekend I sat down and wrote a detailed narrative documentation outlining how, exactly, I built Mr. Chatterbox: www.estragon.news/mr-chatterbo...
This is in the best traditions of DH. Its success didn't depend on big compute or extensive tech background, but on good design + 19c vibes. And yet, in fact, figuring out how to fine-tune something like this is not a solved problem, and people w/ more compute hours could learn from the experiment.
I’ve been overwhelmed and thrilled by the response to this! I’ve also gotten a lot of questions asking how I did it. So, this weekend I sat down and wrote a detailed narrative documentation outlining how, exactly, I built Mr. Chatterbox: www.estragon.news/mr-chatterbo...
I’ve been overwhelmed and thrilled by the response to this! I’ve also gotten a lot of questions asking how I did it. So, this weekend I sat down and wrote a detailed narrative documentation outlining how, exactly, I built Mr. Chatterbox: www.estragon.news/mr-chatterbo...
Mr. Chatterbox, or, The Modern Prometheus
A few days ago, I posted about a personal project that I've been working on for the last few weeks: Mr. Chatterbox, a chatbot trained from scratch on Victorian-era literature. I have to admit, I was t...
estragon.news
Wanted to share a personal project I've been working on: I love classic literature, and I'm also taking an AI class here at UIUC, so I trained an LLM from scratch on Victorian lit from 1837 to 1899. This is Mr. Chatterbox, the Victorian Gentleman Chatbot: huggingface.co/spaces/tvent...
You never know what data will be used for! I uploaded a @britishlibrary.bsky.social dataset to Hugging Face in 2022. IIRC one of my first PR to a HF repo! 4 years later, someone trains a Victorian chatbot on it More libraries should be sharing their public domain collections for AI to build on!
Want to talk to the past? Here' an LLM "trained entirely from scratch on a corpus of over 28,000 Victorian-era British texts published between 1837 & 1899, drawn from a dataset made available by the British Library" Quite different from an LLM roleplaying a Victorian. huggingface.co/spaces/tvent...
Want to talk to the past? Here' an LLM "trained entirely from scratch on a corpus of over 28,000 Victorian-era British texts published between 1837 & 1899, drawn from a dataset made available by the British Library" Quite different from an LLM roleplaying a Victorian. huggingface.co/spaces/tvent...
Guess I found a good use for AI for once. Not enough to make me cease my appeals for Butlerian Jihad, but at least this one's likely to make me laugh a bit.
Wanted to share a personal project I've been working on: I love classic literature, and I'm also taking an AI class here at UIUC, so I trained an LLM from scratch on Victorian lit from 1837 to 1899. This is Mr. Chatterbox, the Victorian Gentleman Chatbot: huggingface.co/spaces/tvent...