Remix.run Logo
▲ ChatGPT for SNES(bxfoundry.com)
60 points by evan_ 7 hours ago | 18 comments
▲soulofmischief 40 minutes ago | parent | next [-]

That's a great project, thanks for sharing.

▲vjay15 2 hours ago | parent | prev | next [-]

this is so cute

▲4lx87 3 hours ago | parent | prev | next [-]

Wild to think we could have implemented transformers in 1990 with consumer hardware. I guess we needed the web first for training, but small transformers are surprisingly effective. What other discoveries are right under our nose?

▲kqr 22 minutes ago | parent | next [-]

As someone else pointed out, training something like this requires much more hardware than running it.

It's similar with other forms of machine learning, too. A logistic regression can make calibrated decisions in complex situations with nearly no compute at all, but training it on lots of data to give it good weights is more expensive. And even if we could submit a training batch job on 1990s hardware and wait a few years for it to complete, there's a level of interactivity required at the start to explore one's way to the right parameters and tokenisation etc. that probably wouldn't have been economically feasible at the time.

In a wider sense, a general purpose computer is a general purpose computer. Hypothetically, we "could have" run transformers in the 1940s on vacuum tubes. It just would have (a) required much space, (b) produced much heat, and (c) required tube replacements all the time. We "could have" run transformers on human computers in the late 1800s or early 1900s – all the maths was there, I think? – but it would have been even slower!

▲phire an hour ago | parent | prev | next [-]

I'm not sure we would have found this impressive back then, the output is pretty nonsensical. Back then, we had other (more manual) approaches that could produce somewhat equivalent results.

It looks impressive to us because we know how good the output from a full-sized LLM can be, and we are willing to give it a lot of leeway because of its small size.

Well... computational linguistics academics should have been impressed at how well it understands language with nothing more than ingesting a huge amount of untagged data.

▲mk89 an hour ago | parent | prev | next [-]

Plenty of research papers out there that get no love.

Go through them and find the next big thing :)

▲LeoPanthera 3 hours ago | parent | prev [-]

Running them isn't the same as creating them.

▲DoesntMatter22 2 hours ago | parent [-]

Sure but it could still have been done, we just didn't realize it.

▲paxy an hour ago | parent [-]

By that logic we could have invented it at any point in history with a turing machine made of rocks.

▲staplehorse6553 an hour ago | parent [-]

And now you're re-reading the Three-body Problem.

▲matt3210 6 hours ago | parent | prev | next [-]

you beat me to it

▲jubilanti 12 minutes ago | parent | prev | next [-]

At this point, it's so small it's just a glorified Markov chain? It produces meaningless English-shaped sentences.

▲lanewinfield an hour ago | parent | prev | next [-]

"How many 'r's are there in the word strawberry?"

> 1. The opposite of strawberries

AGI achieved

▲adiabatichottub 2 hours ago | parent | prev | next [-]

I asked it the answer to the question of life, the universe, and everything. It gave me a nonsensical response. I'd say it works perfectly.

▲teaearlgraycold 15 minutes ago | parent | prev | next [-]

I appreciate that out of a vocabulary of 8178 words, "San" and "Francisco" made the cut.

Also, it has a 128 token context window. Brutal!

▲ronsor 2 hours ago | parent | prev | next [-]

"1 plus 1?"

> The answer is 9.

It tried.

▲redundantly 2 hours ago | parent [-]

Nah, it just used a different meaning for the symbols involved.

▲erickhill an hour ago | parent | prev [-]

Hm, I asked it questions about Metroid and the answers were gibberish. "A couch is a type of fruit" etc.

Eh?