Haeuserschlucht

9 ms·

91.
Ask HN: Do you still make yourself believe some local LLM was helpful?
4 points by Haeuserschlucht 2y ago|3 comments
92.
Ask HN: How do you know whether your prompt is bad or the model sucks?
2 points by Haeuserschlucht 2y ago|1 comments
93.
Deep Learning's Memory Crisis Solved? Introducing the 'Lean Model' Revolution
1 points by Haeuserschlucht 2y ago|0 comments
94.
by Haeuserschlucht 2y ago
> for reading sensitive/non-sensitive docs, contracts Can you elaborate?
95.
by Haeuserschlucht 2y ago
Optimizer detected!
96.
by Haeuserschlucht 2y ago
Great idea!
97.
Ask HN: Disconnected: When the Screen Fades
2 points by Haeuserschlucht 2y ago|0 comments
98.
Ask HN: How do you use local LLMs?
10 points by Haeuserschlucht 2y ago|7 comments
99.
by Haeuserschlucht 2y ago
> The NSA's mission is to spy on foreigners and not Americans. Snowden proved the opposite.
100.
People like to argue that they don't trust deepseek, because it's in China
3 points by Haeuserschlucht 2y ago|2 comments
101.
by Haeuserschlucht 2y ago
Can you explain to me how HIPAA makes sure that this works?
102.
by Haeuserschlucht 2y ago
It's not the price, it's the privacy that drives to local ones.
103.
by Haeuserschlucht 2y ago
Be psyched that it doesn't. Anti-Cheat software is spyware at it's finest.
104.
Local LLMs are much better than their reputation
2 points by Haeuserschlucht 2y ago|4 comments
105.
by Haeuserschlucht 2y ago
Use case?
106.
Burgers by Percent(soloist.ai)
1 points by Haeuserschlucht 2y ago|0 comments
107.
by Haeuserschlucht 2y ago
Sad and disappointing like any dead end station.
108.
Experience VR Gaming Like Never Before(soloist.ai)
1 points by Haeuserschlucht 2y ago|0 comments
109.
by Haeuserschlucht 2y ago
AI dances around actually feasible information when using this prompt.
110.
by Haeuserschlucht 2y ago
Anywhere, it's just for the experience.
111.
by Haeuserschlucht 2y ago
y not just create a filter for "Reply STOP to unsubscribe" and remove them all. Your waste of potential could be used to cure cancer or at least mow the lawn.
112.
by Haeuserschlucht 2y ago
I need to talk to a human
113.
by Haeuserschlucht 2y ago
How does it compare to notebooklm in terms of quality and length of output given llama3.2-8B? You need to give more context, currently it seems like this is yet another reinvention of the wheel.
114.
by Haeuserschlucht 2y ago
I wonder why you use any habit tracker, after all, you notice yourself when you slack on them. And being reminded that you didn't do anything is not helpful and having done a whole month does not guarantee the next month. How do you se
115.
by Haeuserschlucht 2y ago
Your app looks nice and does what was wanted. I wonder why you use any habit tracker, after all, you notice yourself when you slack on them. And being reminded that you didn't do anything is not helpful and having done a whole month do
116.
by Haeuserschlucht 2y ago
Why do you care? Serious question.
117.
by Haeuserschlucht 2y ago
Because mistakes can be easily rectified, IT does not have those strict requirements. Just having to unsolder electronics makes you plan extra tight next time. So the reason for this is that you only get one shot, do not miss your chance to
118.
by Haeuserschlucht 2y ago
What does "I work in VC" mean? I know, venture capital, but what does it mean in practise and how is that tool useful for you?
119.
by Haeuserschlucht 2y ago
Great idea and it works, I am already discouraged and inspired by the results. Did you create this to get inspiration of what you can do that sounds good and has no competition? Be aware, no competition is not always a good sign.
120.
by Haeuserschlucht 2y ago
User: What's the weather in Chicago? Layer1(LLM) to Layer2(AgentWeather): Can you tell the weather in chicago? AgentWeather: I have a tool that's called get_weather. Should I use this? No -> AgentWeather -> LLM: I cannot tel

More ›