HNHacker News
TopNewBestAskShowJobs

mike00632

885 karma · joined April 10, 2018

submissionscomments
mike00632··on GPT-4o
Does anyone else, when writing comments, feel that you need to add a special touch to somehow make it clear that a human wrote it?
mike00632··on GPT-4o
I have never tried GPT-4 because I don't pay for it. I'm really looking forward to GPT-4o being released to free tier users.
mike00632··on GPT-4o
I was shocked at how quickly and naturally they were able to correct the situation.
mike00632··on GPT-4o
For what it's worth, OpenAI also shared videos of failed demos:

https://vimeo.com/945591584

I really value how open they are being about its limitations.

mike00632··on Gemini AI
I've had the same experience. Now ChatGPT is my go-to.

I often paste error messages with context into the chat and ask what's wrong.

mike00632··on LLMs can't self-correct in reasoning tasks, DeepMind study finds
This is a very interesting example. I don't blame LLMs for not understanding which English text makes hard sounds. I wonder if it would get better if it were multi-modal.
mike00632··on LLMs can't self-correct in reasoning tasks, DeepMind study finds
I just asked Chat GPT 3.5 the following: "Please give an exponential equation that has integer solutions which is difficult to solve.", and the response is: "

Certainly, here's an exponential equation that can be challenging to solve for integer solutions:

2^x+3^y=7^z

This equation involves three variables, x, y, and z, and requires finding integer values for these variables that satisfy the equation. This type of equation is known as a Diophantine equation, and solving it can be quite challenging, especially for larger values of x, y, and z.

"

mike00632··on Dude,Where's My Donations? Wikimedia gives another $1M to non-Wikimedia projects
Donate to the Internet Archive/Wayback Machine!

https://help.archive.org/help/category/donations-2/donating-...

mike00632··on LK-99 isn’t a superconductor
Some members of the LK-99 team published a claim of superconductivity and it included falsified data. Nature published it and had to retract it.

https://www.nature.com/articles/s41586-022-05294-9

https://www.science.org/content/article/something-seriously-...

mike00632··on A room-temperature superconductor? New developments
Nature published their last big superconductor claim and then had to retract it. Publications getting burned by these folks makes them understandably reluctant to get worked up again.
mike00632··on Oregon decriminalized hard drugs – early results aren’t encouraging
Doesn't the same apply to alcohol?
mike00632··on A room-temperature superconductor? New developments
What should the headline be? "South Korean scientists claim room temperature superconductivity again! Are they lying like their last paper? Let's celebrate!"
mike00632··on US smartphone shipments fall sharply, but Android more than iPhone
I think this varies by groups of people and is very similar to everyone in the group having Nike shoes. It's a marketing ploy from Apple.
mike00632··on US smartphone shipments fall sharply, but Android more than iPhone
What if we just want to customize our home screen? I use a custom launcher which doesn't require root or a custom rom.
mike00632··on A Senior Engineer’s Guide to FAANG Interviews
The writers who are currently striking have a lot of things to say about Netflix being negative.
mike00632··on Apollo will close down on June 30th
I think they were also considering the crawlers that were using Reddit data to train AI models.
mike00632··on Run Llama 13B with a 6GB graphics card
How much resources are required is directly related to the memory size devoted to each weight. If the weights are stored as 32-bit floating points then each weight is 32 bits which adds up when we are talking about billions of weights. But if the weights are first converted to 16-bit floating point numbers (precise to fewer decimal places) then fewer resources are needed to store and compute the numbers. Research has shown that simply chopping off some of the precision of the weights still yields good AI performance in many cases.

Note too that the numbers are standardized, e.g. floats are defined by IEEE 754 standard. Numbers in this format have specialized hardware to do math with them, so when considering which number format to use it's difficult to get outside of the established ones (foat32, float16, int8).

mike00632··on Google’s quest to digitize troops’ tissue samples
The military members served the public. Google is profiting. You have it backwards.
mike00632··on Google’s quest to digitize troops’ tissue samples
Here is a simple exercise: imagine these were your medical records and your former employer gave them to a third party for profit.
mike00632··on US reps urge DOJ to investigate anticompetitive conduct affecting apartment rent
I don't even have a single house and you're worried about a second house being taxed too much?
mike00632··on Meta lays off 11,000 people
I have very similar thoughts, especially when I compare the Metaverse spots to the products that video game companies make for way less money that are comparably much better. I guess video game companies have a reputation for over-working and under-paying their employees though.
mike00632··on Meta lays off 11,000 people
You criticized Chinese tech companies for being incubated in a walled garden but then criticize TikTok for being subject to US government regulations in the American market? Shouldn't you also give Facebook the same criticism? The only place where Facebook is still experiencing substantial growth is India where their government banned TikTok. It seems to me like TikTok is the real international competitor.
mike00632··on “Science must respect the dignity and rights of all humans”
Yes, exactly. Then you could write popular opinion books about the studies which flesh out the conclusion. Then it could be a motivating tool for an entire political party. You could go on tour to promote your "controversial" books. People will say that the facts must be heard and both sides should be represented.
mike00632··on “Science must respect the dignity and rights of all humans”
This isn't an effort that the Right can use to proclaim racial superiority. Understand the difference?
mike00632··on “Science must respect the dignity and rights of all humans”
You're ASSUMING that race could be linked to "intelligence", that it's important to study, just like I was ASSUMING that handedness might have something to do with spreading COVID. There is clear bias introduced by what we think is important to measure.
mike00632··on “Science must respect the dignity and rights of all humans”
You suggested that the COVID vaccine was somehow bad for being like every other vaccine. I'm suggesting that you need to learn about vaccines and get vaccinated.
mike00632··on “Science must respect the dignity and rights of all humans”
Trans people are literally being murdered. Kids are kicked out of their homes for being trans. The Republican party is currently centering their platform on hatred towards trans people, blaming them for all societal woes. There are now laws being enacted that ban trans people from public places. Yet you not see how this could discourage people from coming out as trans?
mike00632··on “Science must respect the dignity and rights of all humans”
You had to cite an obscure Swedish article to characterize medical care for children as negative? Pathetic.

Why not just trust doctors? Let medical professionals make the decision? How about considering the children you want to hurt instead of your feelings about whether trans people are icky.

mike00632··on “Science must respect the dignity and rights of all humans”
1. Many parents are not fine with trans children. By suggesting that trans children are welcome you are ignoring the very real problems that the Conservative monsters are currently inflicting on them.

2. They benefit from having extra time to make a decision. When you (unnaturally, via the force of law) block children from receiving medical care then you cause them to develop sex characteristics that don't belong to them. If you think about it, you are doing to these children much worse than anything you are suggesting trans advocates are doing.

mike00632··on “Science must respect the dignity and rights of all humans”
Sounds like you're trying to censor information by dismissing the left/right hand hypothesis as "unrealistically stretched".

To put it another way: the way you feel about studying handedness being linked to COVID is the way I feel about studying an assumed link between race and "intelligence". It's not "otherwise interesting" at all. It's ignorant at best.

Page 1 of 24Next →