Viewing profile — fpgaminer

The real nightmare scenario is the AI using its abilities to copy itself to new locations. e.g. hacking into a various cloud services, launching multiple instances of itself, and c…

For large projects like this I think a hierarchical division of labor also helps. If you first carefully define the overall architecture and thus individual high level components o…

If it's a choice between performance and being able to "safely" run sketchy extensions, I'd rather have performance.

Not only that, they additionally ran an experiment with the training temperature turned way up (2.0) and truncation turned off such that the majority of SFT examples were incoheren…

> Is every new thing not just combinations of existing things? If all ideas are recombinations of old ideas, where did the first ideas come from? And wouldn't the complexity of ide…

As far as I can tell 5.2 is the stronger model on paper, but it's been optimized to think less and do less web searches. I daily drive Thinking variants, not Auto or Instant, and u…

I wish they would keep 4.1 around for a bit longer. One of the downsides of the current reasoning based training regimens is a significant decrease in creativity. And chat trained …

Well yeah, because 5.2 is the default and there's no way to change the default. So every time you open up a new chat you either use 5.2 or go out of your way to select something el…

> FPGAs will never rival gpus or TPUs for inference. The main reason is that GPUs aren't really gpus anymore. Yeah. Even for Bitcoin mining GPUs dominated FPGAs. I created the Bitc…

I had to return my Vision Pro after trying it for a week. I'm one of those rare customers that genuinely wanted to keep it, because it's the only VR headset I could _actually_ get …

At least in the U.S. the equality of women in society (and in law) has slowly risen over the last 100 years. Over that same period the availability of pornographic images has also …

"You don't have to be a nice person to be a good person."

Claude will do the same start over if things get too bad. At least I've seen it when its edits went haywire and trashed everything.

I used almost 100% AI to build a SCUMM-like parser, interpreter, and engine ( https://github.com/fpgaminer/scumm-rust ). It was a fun workflow; I could generally focus on my usual …

Maybe this is an insane idea, but ... how about a spider P2P network? At least for local AIs it might not be a terrible idea. Basically a distributed cache of the most common sourc…

Usability/Performance/etc aside, I get such a sense of magic and wonder with the new Agent mode in VSCode. Watching a little AI actually wander around the code and making decisions…

On the vision side of things: I ran my torture test through it, and while it performed "well", about the same level as 4o and o1, it still fails to handle spatial relationships wel…

https://gist.github.com/fpgaminer/8782dd205216ea2afcd3dda29d... That's the model automation. To evaluate the prompts it suggests I have a sample of my dataset with 128 examples. Fo…

I ran an interesting benchmark/experiment yesterday, which did not do Quasar Alpha any favors (from best to worst, score is an average of four runs): "google/gemini-2.5-pro-preview…

Alright, played with it a little bit on the API (Maverick). Vision is much better than Llama 3's vision, so they've done good work there. However its vision is not as SOTA as the b…

So, take a raw LLM, right after pretraining. Give it the bare minimum of instruction tuning so it acts like a chatbot. Now, what will its responses skew towards? Well, it's been pr…

The benchmarks are awful. No disrespect to the people who worked to make them, nothing is easy. But I suggest going through them sometime. For example, I'm currently combing throug…