bogtog

5 ms·

31.
by bogtog 1y ago
I'm using CC right now and I see this: "Tip: Want Claude to remember something? Hit # to add preferences, tools, and instructions to Claude's memory"
32.
by bogtog 1y ago
I agree with the bit about creative writing, and I would add writing more generally. Gemini also allows dumping in >500k tokens of your own writing to give it a sense of your style. The other big use-case I like Gemini for is summarizing
33.
by bogtog 1y ago
> Today, we are releasing updated versions of Gemini 2.5 Flash and 2.5 Flash-Lite, available on Google AI Studio and Vertex AI, aimed at continuing to deliver better quality while also improving the efficiency. Typo in the first sentence
34.
by bogtog 1y ago
Little of this is considered cognitive psychology. The vast majority would be viewed as "social psychology" Setting that aside, among any scientific field I'm aware of, psychology has taken the replication crisis most serious
35.
by bogtog 1y ago
Might just be the channels you're viewing. This video should have a "Jump ahead" appear around 0:19: https://www.youtube.com/watch?v=rs5PfanIkC0&t=1s
36.
by bogtog 1y ago
I opened the subreddit, and the posts/comments I saw didn't show anybody "very irate"...
37.
by bogtog 1y ago
> But my Instagram became TikTok - so now I don't use Instagram. My YouTube is becoming TikTok - so now I don't use YouTube. Everything is implementing autoplaying short form content. X/Twitter. Reddit. etc. There are nota
38.
by bogtog 1y ago
> At the same time, YouTube videos are getting longer, and people are watching more YouTube videos on TVs than on mobile devices I assume this is a replacement for TV/streaming. Cases were you previously would've wanted a 10-mi
39.
by bogtog 1y ago
Would the tax implications not just be for whatever it costs on their end, regardless of what the customer was charged?
40.
by bogtog 1y ago
I assumed all of the labs were already doing something like this to cut down on costs
41.
by bogtog 1y ago
Same here, I can't scroll smoothly at all even when I try (Windows mouse setting set to 15 lines per scroll tick)
42.
by bogtog 1y ago
Same here on all fronts about distractions. I can't tell whether when people talk about listening to music while working/studying: (1) they mean music with no lyrics, (2) they are unserious and okay with their work being constantl
43.
by bogtog 1y ago
If Google agreed to just $34B, getting the cash would be so straightforward that even many random Joes on the street would be able to figure it out
44.
by bogtog 1y ago
> With GPT-5 we will be deprecating all of our prior models Wow, they actually did it
45.
by bogtog 1y ago
When people talk about running a (quantized) medium-sized model on a Mac Mini, what types of latency and throughput times are they talking about? Do they mean like 5 tokens per second or at an actually usable speed?
46.
by bogtog 1y ago
> I highly doubt I am a top 5%er - but wont be shocked if my week ends on a wednessday. I was just starting to use Claude chat more as it is in my subscription but if I can not rely on it to be available for multiple days its functionall
47.
by bogtog 1y ago
Many of these are desirable features!
48.
by bogtog 1y ago
I can't speak to whether it is a parlor trick, but my gut is that processing a 30x30 grid isn't really representative of o3's image processing. This tiny grid isn't like any image it would encounter normally and is so sh
49.
by bogtog 1y ago
I think the top performer afaik (ChatGPT o3) is still treating ARC as a series of characters. I imagine complex reasoning in multimodal processing wouldn't be nearly as advanced so treating it as characters is still better
50.
by bogtog 1y ago
I wonder how much slow progress on ARC can be explained by their visual properties making them easy for humans but hard for LLMs. My impression is that models are pretty bad at interpreting grids of characters. Yesterday, I was trying to ge
51.
by bogtog 1y ago
> If a company invests billions in R&D to create hardware and its integrated software, shouldn’t it have the right to control who or what interacts with it? Why should I be forced to open up the carefully designed ecosystem I’ve buil
52.
by bogtog 1y ago
> If you look objectively, what is Satya's accomplishments? Managing to keep the MS Office grift going and even expand it with MS Teams is something
53.
by bogtog 1y ago
Better yet, "I used machine learning to count all the yurts in Mongolia"
54.
by bogtog 1y ago
I assumed a yurt was a type of person/job, so I initially read the title the same way
55.
by bogtog 1y ago
> vibe coded I'm curious why this needed to be mentioned... none of the visual blurring tools seem to work in a way that I can envision you intended. Also, I sadly suffer from Firefox, which breaks the color-blindness bit
56.
by bogtog 1y ago
That's $2000 but for just 3.5-4.25 tokens/s? I'm hesitant to say that 4 tokens/s is useless, but that is a tremendous downgrade (although perhaps some smaller model would be usable)
57.
by bogtog 1y ago
I think such watermarks are one of DeepMind's goals: https://deepmind.google/science/synthid/
58.
by bogtog 1y ago
> ...and many buildings use millions of gallons of water (often fresh, potable water) per day in their cooling operations. Of note, cooling is water evaporation, so the water will inevitably come back to us as good as new. This contrasts
59.
by bogtog 1y ago
> The largest model we tested has 405 billion parameters, but others, such as DeepSeek, have gone much further, with over 600 billion parameters. Very quickly skimming, I have some trouble taking this post seriously when it omits that th
60.
by bogtog 1y ago
I've also wondered how banks get convinced to offer companies the debt in these LBOs. The only explanation I see is that failures (Toys R Us, etc.) become well-known while successful LBOs and sales of companies with long-term profitabi

More ›