Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
transformi
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
9 ms
·
61.
▲
Ask HN: How training of LLM dedicated to code is different from LLM of “text”
2 points
by
transformi
3y ago
|
3 comments
62.
▲
by
transformi
3y ago
This is very serious. The fact is that they promoted that they not allowing revealed faces in their paper. That's mean that every person that in the training data could be revealed, and there is not transparency about their training d
63.
▲
by
transformi
3y ago
There were many more already week ago... (location & identity restored from trained data..) Causing even more (privacy) concerns.. https://twitter.com/MetaAsAService/status/170679883460343414...
64.
▲
by
transformi
3y ago
I think that the valuable parts are xAI (explanation) and make the interaction between the REAL model and the human.
65.
▲
by
transformi
3y ago
Indeed. Wow what a culture hole! Thanks for the exchange!
66.
▲
by
transformi
3y ago
But why they didn't compare it to SOTA finetuned...(like vicuna playtus..)? ... smells a bit strange..
67.
▲
by
transformi
3y ago
QVC stands for?
68.
▲
by
transformi
3y ago
Looks neat! Do you consider adding input elements? (forms,wizards..) if so it will be my goto platform!
69.
▲
by
transformi
3y ago
Yea that's the first step... but I think that as a group of users we can do much more and build a better internet. This is like Cold-war between companies, that we as users are the cannon fodder...
70.
▲
by
transformi
3y ago
That was escalated quickly. :O Any ideas how we, the users of the internet, should act to prevent those steps? or building a much better connected and not controlled by mega-corporates INTERNET social platform?
71.
▲
by
transformi
3y ago
Related: A similar analogy of 51% attack in crypto to AI-companies https://www.iamnotmj.com/blog/Comparing51ToAI
72.
▲
by
transformi
3y ago
When I think about No-functional approach (like security..), It seems that it is not possible to sell to anyone today...
73.
▲
Ask HN: Trying to figure out AGI side-problem to build startup upon
2 points
by
transformi
3y ago
|
1 comments
74.
▲
Ask HN: Do you think that AI-singularity is the reverse of the big-bang?
4 points
by
transformi
3y ago
|
3 comments
75.
▲
by
transformi
3y ago
Free with installation/ leaving details..... is not FREE....
76.
▲
by
transformi
3y ago
Looks Great tool. :) Suggestion: please add the ability to use credential for social, to be able to perform social tasks.
77.
▲
by
transformi
3y ago
Sounds like onlyfans/ gumroad business model for developers... No doubts some developers will benefits from (like 10%), but it will leave the world less open in my opinion.
78.
▲
by
transformi
3y ago
Great job! Any planning on RTL (right to left?) ?
79.
▲
by
transformi
3y ago
Suggestion: Make it with WEBRTC, instead of phone call...
80.
▲
by
transformi
3y ago
Can LLM help in this task?
81.
▲
by
transformi
3y ago
Yes! You have two options: 1- Find some port for it to run locally on GitHub (it will be quantized and not that useful right now, till 2024-till Qualcomm will ship their phones). - see https://github.com/Bip-Rep/sherpa
82.
▲
by
transformi
3y ago
No thanks. I'll run privately self-hosted llama on my phone. Sharing data with OpenAI it's *irresponsible step* at this time of development.
83.
▲
by
transformi
3y ago
They should focus on the anti-trust connection they made with openai (much more relevant..)
84.
▲
by
transformi
3y ago
Are those pages are generated? (got to say the look authentic but it not that hard to make those...)
85.
▲
by
transformi
3y ago
There are a lot of comment about scraping. But I think that their new standards will try to tag a AI-genererated content. There is not much point to give crawlers a lot of content generation; but rather only the succinct "prompt".
86.
▲
by
transformi
3y ago
SO basically, there is not that functional different...
87.
▲
by
transformi
3y ago
what is the differentiation from RunPod? (they have pretty much same prices but with more services..)
88.
▲
by
transformi
3y ago
same...then i tricked him to give me more credits ;-)
89.
▲
by
transformi
3y ago
Follow-up, anything in graph-theory?
90.
▲
by
transformi
3y ago
Any thoughts about possible applications of this model? Why do they made this source available?
More ›