My man, this is literally what they just did. This isn’t an strawman. Atleast google the meaning of your catchphrase ffs

Something happens Americanly in America

Americans: “What are we, a bunch of üntermench asians???”

It’s not much use with a professional codebase as of now, and I say this as a big proponent of learning FOSS AI to stay ahead of the corpocunts

But ERP is not a cool buzzword, hence it can fuck off we’re in 2025

You’re misunderstanding tool use, the LLM only queries something to be done then the actual system returns the result. You can also summarize the result or something but hallucinations in that workload are remarkably low (however without tuning they can drop important information from the response)

The place where it can hallucinate is generating steps for your natural language query, or the entry stage. That’s why you need to safeguard like your ass depends on it. (Which it does, if your boss is stupid enough)

The model ISN’T outputing the letters individually, binary models (as I mentioned) do; not transformers.

The model output is more like Strawberry <S-T-R><A-W-B>

<S-T-R-A-W-B><E-R-R>

<S-T-R-A-W-B-E-R-R-Y>

Tokens can be a letter, part of a word, any single lexeme, any word, or even multiple words (“let be”)

Okay I did a shit job demonstrating the time axis. The model doesn’t know the underlying letters of the previous tokens and this processes is going forward in time

No, this literally is the explanation. The model understands the concept of “Strawberry”, It can output from the model (and that itself is very complicated) in English as Strawberry, jn Persian as توت فرنگی and so on.

But the model does not understand how many Rs exist in Strawberry or how many ت exist in توت فرنگی

Broadcom management deserve gulag

For usage like that you’d wire an LLM into a tool use workflow with whatever accounting software you have. The LLM would make queries to the rigid, non-hallucinating accounting system.

I still don’t think it would be anywhere close to a good idea because you’d need a lot of safeguards and also fuck your accounting and you’ll have some unpleasant meetings with the local equivalent of the IRS.

This is because auto regressive LLMs work on high level “Tokens”. There are LLM experiments which can access byte information, to correctly answer such questions.

Also, they don’t want to support you omegalul do you really think call centers are hired to give a fuck about you? this is intentional

Yeah this is on me, I’ve never done web dev in my life. Always been low level shit for me. (The project is a highly networked system with isolates, async/await, futures, etc)

But damn man comparatively JS/TS seems a lot easier ¯\_ (ツ) _/¯

Honestly? They’re based for being so easy to make

For the record, I am a C/Dart/Rust native dev 2+ years deep in a pretty big project full of highly async code. This shit would’ve been done a year ago if the stack was web based instead of 100% native code

The infrastructure can be duplicated and studied tho. Would be cool if a little dreamy

Watch someone reverse the thing into turbocharged WINE

Industrial emulation is easy to do, a sandboxed and controlled VM won’t die from hardware faults like a hunk of shit from 1993

Also there are NEW computers made specifically for this particular purpose, they even have ISA buses and shit

I don’t understand why lemmy is living in la la land, the moment you go against the narrative you’re brigaded to shit

Yes, y’all do be in fact wrong

Bonus: IBM sells emulation packages for migration to new architectures. IBM probably knows better than the lot of us.

Removed by mod

Instead of using old proprietary shit you could use Linux or *BSD with a vintage desktop environment and have a blast

Something I noticed is that basic users (someone using a fucking 30 y/o OS is definitely one) have an easier time with *nix because most “technical” people are overfitted and brainwashed to the Micro$uck ecosystem

Meta’s recent LLAMA models are a disaster and worse they only masquerade as open models. Meanwhile Europe has it’s own AI research centers like Mistral who make really good models under the Apache 2 license.

We got baited by piece of shit journos

It’s a local model. It doesn’t send data over.

If they want call data they can buy it straight from service providers anyway

deleted by creator