Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

My 2.5 year old laptop can write Space Invaders in JavaScript now (GLM-4.5 Air)

simonwillison.net

Post reply on HN
Print view View original

411–415 of 415 posts

Re: My 2.5 year old laptop can write Space Invaders in JavaScript now (GLM-4.5 Air)

#411

Post by Breza »

bob1029 wrote: post #211

> still think it’s noteworthy that a model running on my 2.5 year old laptop (a 64GB MacBook Pro M2) is able to produce code like this—especially code that worked first time with no further edits needed. I believe we are vastly underestimating what our existing hardware is capable of in this space. I worry that narratives like the bitter lesson and the efficient compute frontier are pushing a lot of brilliant minds a…

Very well put. There's a lot to be gained from using smaller models and existing hardware. So many enterprise PMs skip straight to using a cutting edge LLM via API. There are many tasks where a self-hosted LLM or even a finetuned small language model can either complete a preliminary step or even handle the full task for much less money. And if a self-hosted model can do the job today, imagine what you'll be able to do in a year or five when you have more powerful hardware and even better models.

Re: My 2.5 year old laptop can write Space Invaders in JavaScript now (GLM-4.5 Air)

#412

Post by Breza »

vessenes wrote: post #303
photon_lines wrote: post #201

Earlier quoted context omitted.

This is why I love using the Deep-Seek chain of reason output ... I can actually go through and read what it's 'thinking' to validate whether it's basing its solution on valid facts / assumptions. Either way thanks for all of your valuable write-ups on these models I really appreciate them Simon!

Nota bene - there is a fair amount of research that indicates models outputs and ‘thoughts’ do not necessarily align with their chain of reasoning output. You can validate this pretty easily by asking some logic or coding questions: you will likely note that a final output is not necessarily the logical output of the end of the thinking; sometimes significantly orthogonal to it, or returning to reasoning in the middl…

That's a good note. I use DeepSeek for early planning of a project because of how valuable its reasoning output can be. It's common that I'll describe my problem and first draft architecture and see something in the output like "Since this has to be mobile optimized..." Then I'll stop generation, edit the original prompt to specify that I don't have to worry about mobile, and run it again.

Re: My 2.5 year old laptop can write Space Invaders in JavaScript now (GLM-4.5 Air)

#413

Post by 0x457 »

pbhjpbhj wrote: post #394
0x457 wrote: post #286

Earlier quoted context omitted.

Because internally, there isn't a difference between correctly "recalled" token and incorrectly (hallucinated).

Depends on the training? If there was eg RLHF then those connections are stronger and more likely; that's a difference (but not a category difference).

Yes, but I thought we're talking about category difference.

Proper RLHF surely boosts "predicted next token until it couldn't" to feel more like "actually recalled".

Re: My 2.5 year old laptop can write Space Invaders in JavaScript now (GLM-4.5 Air)

#414

Post by Breza »

DonHopkins wrote: post #376
xianshou wrote: post #282

I initially read the title as "My 2.5 year old can write Space Invaders in JavaScript now (GLM-4.5 Air)." Though I suppose, given a few years, that may also be true!

Given a few years your 2.5 year old will be a 5.5 year old, too!

Ugh don't remind me. My daughter's fifth birthday is tomorrow and with how fast she's growing I feel like her 15th is on Thursday.

Re: My 2.5 year old laptop can write Space Invaders in JavaScript now (GLM-4.5 Air)

#415

Post by seunosewa »

arkmm wrote: post #224
techwizrd wrote: post #143

Earlier quoted context omitted.

We have been fine-tuning models using Axolotl and Unsloth, with a slight preference for Axolotl. Check out the docs [0] and fine-tune or quantize your first model. There is a lot to be learned in this space, but it's exciting. 0: https://axolotl.ai/ and https://docs.axolotl.ai/

When do you think fine tuning is worth it over prompt engineering a base model? I imagine with the finetunes you have to worry about self-hosting, model utilization, and then also retraining the model as new base models come out. I'm curious under what circumstances you've found that the benefits outweigh the downsides.

When prompt engineering isn't giving you reliable results.
Post reply on HN