Occasional biologist, and a friend to plants everywhere. she/her.
Sooke, B.C, Canada
Joined July 2007
- Tweets1.2K
- Following1.7K
- Followers790
- Likes8.2K
Heads up if you are trying out gemini as a swap in for an openai api call it will fail with api error or list index out of range if it tries to output OpenAI or ChatGPT.
Our discord bot is struggling since brain transplant because of it.
Chris Tuttle retweeted
its almost like.. embeddings was compression the whole time.
wordcount is a embedding and a compression that loses position sensitive to common words
tdidf is an embedding and its a compression that is insensitive to document frequency
gzip is an embedding like tfidf
but contains positional information and a has a tokenizer.
Absolutely love this.
this paper's nuts. for sentence classification on out-of-domain datasets, all neural (Transformer or not) approaches lose to good old kNN on representations generated by.... gzip aclanthology.org/2023.findin…
Chris Tuttle retweeted
The fact that Instant Pot is already being framed as a corporate cautionary tale—the company that went bankrupt bc they made a product so durable & versatile that its customers had little need to buy another one—instead of as a critique of capitalism is deeply, deeply depressing.
Chris Tuttle retweeted
Just thinking of inverting another AI trope--a world where AI becomes increasingly dangerous to humans if it doesn't become sentient or self-aware. Consciousness turns out to be the catalyst for alignment.
Pretty sure a lot of new ‘scraping’ is actually a person getting an AI to browse a single site for them at a time, not an automated crawl of everything. Smart companies should also have a AI-friendly page, simple and with all text in one place, if they want to stay relevant.
Easily solution: Allow scraping, be open like the internet was meant to be. Wikipedia surely is less funded than this place and allows all forms of scraping known to man.
3
4
706
Chris Tuttle retweeted
I'm fed up. Something has to be done about the public discourse on AI. The doomers are spending totally insane amounts of money spreading their message while pretending to be meek underdogs. Time to get organized, even if that requires I do some of the organizing.
Chris Tuttle retweeted
Introducing 🥁🥁...
The WombatWhiz🔥
✨We have just added our own ChatGPT/AI in our Discord community - its name is WombatWhiz.
🪄Join our discord and try it out!
discord.gg/wombatgamers
Still waiting for OpenAI’s Code interpreter? Feel free to play with the appgen button in Webby webby.christuttle1.repl.co/ free to play with unless/until we get too hammered API wise. I’d love any feedback!
Chris Tuttle retweeted
Our first neurogenetic study of ants survived peer review and just appeared OA in its final form @CellCellPress @CellPressNews: cell.com/cell/fulltext/S0092…. @teraxurato & team @RockefellerUniv created the first transgenic ants and identified a sensory hub for alarm behavior. 1/8
AAAH GPT 3.5 16k is so fast and it is so liberating to up all my token limits. webby.christuttle1.repl.co/ is updated now, so quick to just change my limits and add -16K to the api call. Good job @OpenAI.
P.S appgen is silly now, you can get a working 10K token app in one go
71
16k 3.5 works and is Faaast at least right now. casually doubling pretty much all the token limits in my projects lol
lots of good updates including function calling and 16k context 3.5-turbo: openai.com/blog/function-cal…
72
Chris Tuttle retweeted
openai.com/blog/function-cal…
-new function calling capability in the API
-updated & more steerable versions of gpt-4 & gpt-3.5-turbo
-new 16k context window of gpt-3.5-turbo
-75% cost reduction on our v2 embeddings model
-25% cost reduction on input tokens for gpt-3.5-turbo
One of the things I’ve been working on is a little workspace/remixer using GPT 3.5 API. One feature is the ability to run code written by the AI instantly. Lots of stuff works. Plus the AI can use GPT 3.5 api in the apps it makes. webby.christuttle1.repl.co/ WIP
1
115