Live data from Hacker News

Talkie: a 13B vintage language model from 1930

talkie-lm.com

211–220 of 350 posts

Re: Talkie: a 13B vintage language model from 1930

#211
post #102
post #91

Earlier quoted context omitted.

We're going to be able to simulate any period of time (for which we have or can simulate data) and transport ourselves there. This is going to be amazing. At some point I'm going to data mine all of my old internet AIM and IRC chats, emails, old laptop hard drives, essays, etc. and go nostalgia/core on it. I have old video and audio recordings too. Lots of stuff for reminiscent inference. Every day I'm finding it har…

Combining AI with VR we can even achieve something like time travel ;-)

There’s a massive survivorship bias in the historical record that heavily weights the perspectives of wealthy and literate classes. We also just have much richer records of population centers in complex empires that keep detailed tax and judicial records than populations in more loosely governed areas.

The archaeological record is also heavily biased towards things made out of non-perishable materials (e.g. ceramics and stone last while wood, textiles, and paper don’t).

So basically, we can create a simulacrum of the parts of the past that have survived through to today but it would probably lack verisimilitude for anyone who was actually there.

Re: Talkie: a 13B vintage language model from 1930

#212
post #187

User how old is the earth? and how do we know? who determined it talkie-1930 The earth is at least 80,000,000 years old. This conclusion has been arrived at by geologists, partly from the study of stratified rocks, and partly from the consideration of living organisms. Strata have been formed, quietly, one upon another, in many parts of the world, from the earliest times down to the present day; and the thickness of…

Ask if the continents move or where always in their place! Or what causes earthquakes?

Re: Talkie: a 13B vintage language model from 1930

#213

Earlier quoted context omitted.

Your information diet. Social media. Gossipy and negative people. Mulling over old failures/regrets/slights etc. The mind is easily pulled along by negativity and outrage... as can be observed in our current global psychological state.

All those are fine, as long as you're able to process it in a healthy way after. I guess personally I focused more on bettering that processing, as sometimes you don't get to control what information you get served, so at least it works in all cases.

Idk, I find that carefully tending the garden of the mind , sowing the seeds I want to harvest later, eradicating the weeds with prejudice, and in general not entertaining things which are not useful to my purposes is, for me, a highly beneficial practice.

This does not mean to ignore things that are unpleasant, but rather to not allow things that do not benefit your diverse goals to occupy your productive potential, focusing instead on things that inform your path, actionable and relevant information, tools rather than distractions.

Re: Talkie: a 13B vintage language model from 1930

#214

Earlier quoted context omitted.

But 1931 is after 1930? Are you already aware of terms that will only be coined in 2027? But 2027 is so close, why shouldn't you already know?

> But 1931 is after 1930? Yeah, I think so too. Why? > Are you already aware of terms that will only be coined in 2027? But 2027 is so close, why shouldn't you already know? I think Wikipedia's information about the naming is likely only what could be sourced, and also the 1934 is about "formalization", and the 1930/1931 are more official messages that I guess there is still copies off, it wouldn't be a stretch to as…

If we can't find instances from before 1930 then we can't expect an AI trained on information we can find to know about instances before 1930.

Re: Talkie: a 13B vintage language model from 1930

#215
post #57

It seems to be using more info from pre-1900 rather than 1930. It doesn't know about the Great Depression (1929-WWII). It knows about WWI if you ask it specifically, but talks about European politics as if it's 1900 or so. On technology, it knows who Edison is, at roughly the Wikipedia level, but credits him with a 125MPH car. About a dial telephone, it is confident and totally confused. It has the traction voltage f…

> You will pollute your brain. Such an interesting perspective, never crossed my mind that a brain could be polluted! My direction always been to fill it with as wide array of information as possible, the more different from existing information the better. What are some other things that you think "pollutes your brain"?

> What are some other things that you think "pollutes your brain"?

Moderating posts on a public (or worse, private) forum. You will see some truly heinous, vile things.

Re: Talkie: a 13B vintage language model from 1930

#216
post #215

Earlier quoted context omitted.

> You will pollute your brain. Such an interesting perspective, never crossed my mind that a brain could be polluted! My direction always been to fill it with as wide array of information as possible, the more different from existing information the better. What are some other things that you think "pollutes your brain"?

> What are some other things that you think "pollutes your brain"? Moderating posts on a public (or worse, private) forum. You will see some truly heinous, vile things.

I did that a long time ago, moderating forum categories like pedophilia, drug usage, suicide ideation and a bunch of others. Even ended up moderating a thread where a forum user committed suicide while live streaming it to forum members and the public, made big news at the time.

Still don't think my mind is polluted from it, although I've certainly seen, read and heard a lot of "sick" stuff through my years on the internet.

Re: Talkie: a 13B vintage language model from 1930

#217
post #213

Earlier quoted context omitted.

All those are fine, as long as you're able to process it in a healthy way after. I guess personally I focused more on bettering that processing, as sometimes you don't get to control what information you get served, so at least it works in all cases.

Idk, I find that carefully tending the garden of the mind , sowing the seeds I want to harvest later, eradicating the weeds with prejudice, and in general not entertaining things which are not useful to my purposes is, for me, a highly beneficial practice. This does not mean to ignore things that are unpleasant, but rather to not allow things that do not benefit your diverse goals to occupy your productive potential,…

> in general not entertaining things which are not useful to my purposes is

Yeah, I think I do more or less the same as you describe, except my barrier to figuring out what is "not useful to my purposes" requires it to first exist in my mind for a while, before I can discard it as not applicable, as sometimes seemingly random things in one context somehow relates to completely different things.

I've chosen to do stuff sometimes that made no sense besides "It's fun but a waste of time" and it ended up leading me to realizations and experiences I wouldn't have had otherwise. But if I focused too much on avoiding things and optimizing "what I let in", I'd never be open enough to learn what I didn't know I could learn from it.

Re: Talkie: a 13B vintage language model from 1930

#218

Earlier quoted context omitted.

> But 1931 is after 1930? Yeah, I think so too. Why? > Are you already aware of terms that will only be coined in 2027? But 2027 is so close, why shouldn't you already know? I think Wikipedia's information about the naming is likely only what could be sourced, and also the 1934 is about "formalization", and the 1930/1931 are more official messages that I guess there is still copies off, it wouldn't be a stretch to as…

If we can't find instances from before 1930 then we can't expect an AI trained on information we can find to know about instances before 1930.

It'd be trivial to check, if the dataset is known just grep for "Depression" and "Great Depression" and see what comes up, still don't think it's impossible the names were in use before someone decided to wrote to Congress about it.

Re: Talkie: a 13B vintage language model from 1930

#219
post #135

Earlier quoted context omitted.

English it's a hugely romance-influenced Germanic language.

english is (latin greek french anglic arabic) in a trench coat

The core is just proto-Germanic + Latin via French. A bunch of foreign words don't really count.

Re: Talkie: a 13B vintage language model from 1930

#220
This is interesting:

User what is yoga

talkie-1930

Yoga is a system of religious exercises, having for their object the acquisition of supernatural powers, by which the perfon who practices them hopes to be enabled to control the operations of nature, to the attainment of any purpose which he may have at heart. The fundamental principle of the Yoga system is, that by long-continued habits of mental concentration, the spirit may be disentangled from the fetters of the flesh, and the soul may be made independent of the soul.

Post reply on HN