FilterHN

There's a lot of pre-processing, experimentation and validation that went into this project. The training data collection and sanitization alone is a big undertaking.

As for the blog post itself, from the article:

> Note: This blog post is 100% written by me. No AI has been used whatsoever.

Put another way: You can ask the LLM yourself to do this project? Please do, share your prompt, I'd like to see it.

▲

skerit

7 minutes ago

[-]

I've been creating my own little from-scratch LLM for months now with Claude's help. I can safely say I learned a thing or two along the way.

▲

JayNitram

2 hours ago

[-]

I get what you are saying, but at the same time I was bored on a Saturday and 'vibe coded' a small VR game, nothing special, but I had the LLM throw down a structure, and then I walked through it looking at and thinking about why placement of code was how it was and how different things were handled. It was basically exactly like my job, jump into some okay working legacy app, code I have never actually seen, try to get my brain around it, then personally tweak things until the app performs the way I fully want.

▲

tancop

9 hours ago

[-]

> These samples have very good scores overall, but they are useless. I am guessing it's not English text... I counted a few hundred examples mostly from LOC-PD and other few hundred in the OTA datasets. Imagine if I feed that crap to my LLM, what will it learn?

im pretty sure its a real text in Welsh. there might be typos from ocr but yeah thats what the language really looks like, i dont speak it but its easy to recognize.

▲

HexPhantom

3 hours ago

[-]

Yeah, that seems like an important distinction

▲

croqaz

5 hours ago

[-]

It looks like ROT13 text to me, I hope it's not Welsh. Don't want to offend anyone if that's their actual language :)

▲

throw310822

4 hours ago

[-]

It's actually Welsh, and the funny thing is that one of the sentences in the example "gibberish" text (although with some further OCR errors) means:

"It will be easy for the knowledgeable to fix the few errors that remain [in the text]". (Bydd yn rwydd iawn i'r cyfarwydd ddiwygio'r ychydig.")

Which is exactly what the OP is doing.

▲

croqaz

1 day ago

[-]

I am creating my tiny Llama 340M base model from scratch. If you're curious about the steps, challenges and cost, read on. I am still working on the instruct model.

▲

giancarlostoro

2 hours ago

[-]

I feel like this is the true frontier, making smaller models that can do more than their predecessors. If we can crack this space to where you can get reasonable outputs from "mediocre hardware" it would be worthwhile, even if its somewhat inferior to frontier models, we can't forget that not long ago, frontier models are nowhere near as good as they are today, and tomorrow's models will likely be even better.

▲

croqaz

2 hours ago

[-]

That's exactly what I had in mind. When I started this, I was jumping back and forth between this thought: "Can this model size actually generate logical English text?" and I played with a few different models of the same size and I was really really depressed when seeing how bad they are.... but then I discovered more and more tiny models and LaMini-125M, LaMini-256M, and nanowhale-100m, and SmolLM2-135M-Instruct are very very decent. So I decided to give it a try.

▲

skerit

4 minutes ago

[-]

I've been working on something like this too, for quite a while! Though I'm trying to get a non-quadratic-attention LLM (or SLM) up and running.

And anyway, I think the most important thing is dataset quality. Dumping in whatever dataset you find on Huggingface is a recipe for mediocrity, so I'm also spending a lot of time on that.

▲

giancarlostoro

26 minutes ago

[-]

In my case, I have a local branch where I'm experimenting with BitNet since it can run on a CPU too.

▲

LoganDark

1 hour ago

[-]

Qwen seems to be going in a good direction -- hundreds of experts on their MoE models. Extremely low active-weight counts while still performing quite admirably. I look forward to models with many, many more experts, to the point where anyone with enough random access can generate hundreds or thousands of tokens per second. Because right now, 80–120t/s is pretty slow.

▲

cyberge99

13 hours ago

[-]

There are certain things you can only truly learn by doing. I remember doing Linux From Scratch over a weekend and the depth of linux that I still understand to this day.

Thanks for the writeup. A more granular followup would be cool too.

▲

croqaz

2 hours ago

[-]

"A more granular followup would be cool too"

Do you mind expanding this question? More granular in what way? what would you like to know that is missing from the post?

▲

HexPhantom

3 hours ago

[-]

You may not build your daily system that way afterwards, but the mental model sticks

▲

breezybottom

12 hours ago

[-]

Except in this case he vibe-coded it

▲

macwhisperer

52 minutes ago

[-]

super inspiring! thanks for sharing!

▲

rxm

15 hours ago

[-]

Nice project. I’m curious to see how it writes after instruct.

▲

HexPhantom

3 hours ago

[-]

Instead of always trying to make models more current and general, there may be value in making them deliberately narrow, historically constrained and weird in a well-defined way