back
user profile
padolsey
4,466karma·578submissions·June 29, 2011
about
padolsey.at.hn
I'm James. Living between Beijing and London. I like coding. Also plants. Stroke survivor & disability advocate. My dog's a whippet/iggy cross and is called Ducky. He's a beautiful lunatic.
* website: [j11y.io](https://j11y.io) * building: [nope](https://nope.net) * bsky: [@j11y.io](https://bsky.app/profile/j11y.io) * contact: https://tally.so/r/waYPvE * twitter: [@padolsey](https://x.com/padolsey) * book recommendations: [ablf.io](https://ablf.io) * me = founding eng @ [collective intelligence project](https://cip.org)
My work email is my first name at cip dot org.
---
---
My meet.hn token: meet.hn-10d164d4-bf7f-4431-898f-e02b34b836dc
recent activity (578 total)
comment
The follow-up's better imho: https://digitopoly.org/2024/03/07/after-the-gold-rush/ E.g. On biases in LLM responses and lack of explainability: > Believe i…
comment
I don’t think you’d be fine-tuning a whole model in such cases. That seems over the top, no? I assume you’d get sufficiently far with big context windows, vector search, RAG. Etc.
comment
Ooh intriguing. I’m learning mandarin at the moment, and also happen to have some sodium valproate lying around for my epilepsy. I’m not currently taking it but I could give it a go and report back.. …
comment
> When it came to light that governments were segregating children in the same age group in school, making them sit in separate classrooms, and even teaching them from different textbooks – said to…
comment
Like others, I think the categories are a bit iffy. But tbh It's really hard to correctly delineate stuff. Terms like 'IDE', 'workspace', 'pipeline', 'workflow&…
comment
Hmm well I believe reddit made up a huge portion of the training data for GPT2 but yes, tbh I have no support for the claim that that's the case with current versions. Anyway, I guess if we consi…
comment
Considering its corpus, to me it makes almost no sense for it to be more helpful when offered a tip. One must imagine the conversation like a forum thread, since that’s the type of internet content GP…
comment
Raisin: really easy mobile app: https://www.raisin.co.uk/ Flagstone: 10k minimum, old-school but well-trusted, great rates for higher deposits: https://www.flagstoneim.com&…
comment
Yes to that. And this may be my own risk-averseness, but I don't have complete confidence in all these derivative instruments anyway. I don't have time or sufficient interest to look into th…
comment
Part of that diversification is the money markets, no?
comment
Oh indeed. I believe these models should be uncensored. But as we’ve seen with the latest SD model, and with the much more locked down ‘LLM-fronted’ image generators from OpenAI and Google, safety is …
comment
This is so quick! The demo being publicly consumable is powerful, though SDXL is not immune to abuse or NSFW (and possibly illegal) generations. I wonder who ends up being held accountable for such ge…
comment
Check out https://scenario.gg - they let you train your own LoRAs on custom images of a character (you need around 20 or so images from different angles for good consistency). A bit simple…
comment
I’ve been thinking about the anger response recently as me and my partner are thinking about becoming parents. Some of our friends with children seem to be liable to anger, and it is quite a stereotyp…
comment
I definitely see it as self-serving too, yes, but I also see it as a convenient temporary alignment of incentives. The world and its regulators definitely need time to adjust and educate themselves, s…
comment
With everything we've seen in the last couple years, do you sincerely believe that all of those points won't be solved pretty soon? There are many intermediary models that can be used to rem…
comment
"The world doesn't work that way". Quite pessimistic a position to hold here, no? We–in technology especially–are in positions of significant leverage. We should be talking about how we…
comment
Nah just fine-tune the model to a specific set of characters or aesthetic. It's not hard, already done with SDXL LoRAs. You can definitely generate a whole movie from just a storyboard.. if not n…
comment
Hmm who's this for? vast.ai seems far more reputable and competitive: https://cloud.vast.ai/create/ Why would I bother getting in touch with a random seller when there'…
comment
On TaoBao here in Beijing I can get one for around 95USD…. No strong indication whether it’ll work tho! https://m.tb.cn/h.5Hbc2x0sHhXTXZG?tk=nl7YW9gh6MR …
comment
Very problematic. How does the vehicle know which pole to talk to? How does it know it’s not incorrectly communicating with a pole on a parallel or perpendicular road, for example? Or indeed a road th…
comment
Yeh. Makes it intimidating to try to make a startup until you realise it’s just generic polish over some cobbled-together SaaS APIs and docker containers. That’s what depresses me. You don’t need good…
comment
The article seems to suggest that the role of crypto to these young Chinese individuals is that it forms a means of community and a nonjudgemental channel of new thoughts. I feel it's become quit…
comment
So, best to do it without publicizing. :D
comment
> Every tech company minus the few doing core research have been doing this for at least half a year I'm assuming you mean all those new 'AI wrapper' startups popping up.? I wouldn&#…