Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
lostmsu
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
34 ms
·
811.
▲
by
lostmsu
10mo ago
Did you train the model with quantization awareness? How?
812.
▲
by
lostmsu
10mo ago
So many words with 0 useful information.
813.
▲
by
lostmsu
10mo ago
Does this include Gemini CLI the terminal vibe coding?
814.
▲
by
lostmsu
10mo ago
None of these answers are correct btw.
815.
▲
by
lostmsu
10mo ago
How does it go with your wife when opinions are opposed and the matter is important (such as kids)?
816.
▲
by
lostmsu
10mo ago
You're behind by 1.5 years on that thought. They certainly can.
817.
▲
by
lostmsu
10mo ago
If you have latency detection already why not pause H.264 frames, then when ack comes just force a key frame and resume (perhaps with adjusted target bitrate)?
818.
▲
by
lostmsu
10mo ago
This picture is wrong because deaths themselves are not as interesting as estimated years of live lost. For cancer and heart disease that multiplier would be around 5x (e.g. mostly older people die from cancer, but if they did not, in ~5 ye
819.
▲
by
lostmsu
10mo ago
I did the same with recent stuff and so far gpt-oss-120b on high was the best with gpt-oss-20b on high close second.
820.
▲
by
lostmsu
10mo ago
Are you sure about OpenAI? I thought they actually do retain your agent chats (training I am less concerned about personally). Anthropic has an option to opt out of training and delete the chats from their cloud in 30 days.
821.
▲
by
lostmsu
10mo ago
I bet on km/h vs mph mistake.
822.
▲
by
lostmsu
10mo ago
This makes me think: are there any "pair-programming" vibecoding tools that would use two different models and have them check each other?
823.
▲
by
lostmsu
10mo ago
Both your examples require Internet access so there's no reason not to use cloud-hosted model which would work magnitudes better.
824.
▲
by
lostmsu
10mo ago
Would have to be ChatGPT these days.
825.
▲
by
lostmsu
10mo ago
Last I checked the Turing test stands. I've only seen reports of LLMs winning under some weird conditions. Interestingly, these were a year or two ago, and nobody seem to have tried Turing tests lately with newer LLMs.
826.
▲
by
lostmsu
10mo ago
Yeah, if you had experience with their Jetson boards, you'd know Nvidia is not well regarded for their OS support. TK1 support stopped after under 4 years. Basically they released it with some version of Ubuntu (14.04 LTS) and never up
827.
▲
by
lostmsu
10mo ago
Compute demand likely improved today's availability and performance of hardware.
828.
▲
by
lostmsu
10mo ago
Report to police anonymously and have them stopped might be an option. If you can't convince them the money might.
829.
▲
by
lostmsu
10mo ago
Now compare batched training performance. Or batched inference. Of course prefill is going to be GPU bound. You only send a few thousand bytes to it, and don't really ask to return much. But after prefill is done, unless you use batche
830.
▲
by
lostmsu
10mo ago
in Rust
831.
▲
by
lostmsu
10mo ago
> 3090 would be nice They would need 3x speedup over the current generation to approach 3090. A100 that has +- the 3090 compute but 80GB VRAM (so fits LLaMA 70B) does prefill at 550tok/s on a single GPU: https://www.reddi
832.
▲
by
lostmsu
10mo ago
It's further away so it's pulled less, so it shifts less.
833.
▲
by
lostmsu
11mo ago
The consensus is that nobody should have root SSH to a production server.
834.
▲
by
lostmsu
11mo ago
Why do you think it's Claude and not iTerm?
835.
▲
by
lostmsu
11mo ago
You were renting and had a job, then had a bad trip that crushed your intelligence/mental health, causing you to get laid off and evicted.
836.
▲
by
lostmsu
11mo ago
Fully utilizing hardware is not trivial.
837.
▲
by
lostmsu
11mo ago
> One second ... unbearable pain, ... next second ... no pain This contradicts "No it doesn't".
838.
▲
by
lostmsu
11mo ago
I don't see how lack of either of the suggested options is any kind of indication that the person doesn't accept that success has a factor of chance. I doubt there's even a claim that this is right ethically or that you are n
839.
▲
by
lostmsu
11mo ago
That heritability doesn't cover all genetic factors. E.g. out of 100% of IQ variation 50% might be inherited, but that doesn't say that the rest is nurture, right? It can still have a huge factor of genetic lottery. E.g. isn'
840.
▲
by
lostmsu
11mo ago
The author of this piece totally ignored that heritability is only part of the genetic lottery.
More ›