SamInTheShell

How much does it actually matter? I use to bother with these settings, but never really seems to fix the problems I commonly see. The issues I commonly see are small models getting stuck in a repeating loop, or they just fail at follow up calls for editing.

Build a distributed system using raft groups. You will see Opus fail.

Opus can make harnesses fine. I barely have to instruct it.

This is probably the first small model I got through some simple web game tests without having to reset the context. It tends to opt to overwrite an entire file instead of doing edits... which editing is where most of these small models fall apart along with getting stuck in repeating loops. Only 24k tokens in so far, it did some decent newbie work.

Tool calling at Q8 for this model is kinda impressive. It's doing patch edits like a champ. It gets stuck looping and doing the wrong things though as the context window grows.

Related/Dupe: https://news.ycombinator.com/item?id=48494697

Also, I had Claude Opus read through some of the mailing list threads and it made this script to check a few things: https://gist.github.com/SamInTheShell/c3609743d742f002f1e8c7...

Lazy check script by Claude Opus: https://gist.github.com/SamInTheShell/c3609743d742f002f1e8c7...

It seems my stubborn resistance to using yay saved my bacon here.

Good luck everybody else.

Scratching my head. Wondering why I would reach for this over just running a Yugabyte cluster.

It’s just stuff I play with. My main reason in publishing this is to give people that otherwise lack the knowledge can have somewhere to start.

This should be safer than relying on seatbelt on Mac OS. Docker and Podman being in a VM is pretty much peak security imo for AI.

Showed me teal but only have options for blue and green. Teal ain't either.

Edited: I saw the multiplier table changes for annual subscribers. Not happy about it and planning to demand a pro-rated refund.

I don’t even read the conversational garbage. Only care about the outputs I demanded it give me.

I actually changed this tonight. Most of my feedback so far has been this issue, so I just changed it so the ship fires where the mouse is, movement is still WASD.

This was a small side project I worked on this last week. All the code is AI generated.

The actual link from the initial reddit post without the GOOG tracking: https://www.congress.gov/bill/119th-congress/senate-bill/136...

Edit: Also does this mean OpenAI can bring back the Sky voice?

Yeah, I plan to do a follow up comment with data and results.

Thought about this during the morning. I'm run the posts through ministral3:3b, mistral-small3.2:24b, and gpt-oss:20b this weekend to build a sentiment mapping and see what I get. I'm optimistic about ministral3:3b, but the other two are pretty good at this type of stuff.

To me this seems to be lacking in usefulness. Maybe if I was hard "no AI", it might look useful, but how do you trust people anymore? You can't really follow the "trust but verify" philosophy here.

The only way to really have a GenAI free codebase today is simply by not accepting any external contributions.

Between banking, infra, or government institutions, you've already got a relationship with a vibe coder. You can't avoid it unfortunately.

It is only funny until that vibe coder is building the data warehouse that holds your data and doesn’t catch the vulnerability that leads to your data leaking.

Perhaps I can laugh at the next Equifax of the world as my credit score gets torched and some dude from {insert location} uses my details to defraud some other party. Of which I don’t find out about until some debt collector shows up months later.

Not surprised at all. Rage bait has been leading to clicks since before the dawn of AOL.

About that data though, just publish that. Throw the data and tooling up on github or huggingface if it's a massive dataset. Would be interested in comparing methodologies for deriving sentiment.

Low tech means accessible. A cowboy hat works better.

I’d urge you to report it anyway. As someone that does use these tools I’m always on the lookout for other people pointing this type of stuff out. Like the .claude directory usage does irk me. Also the concise telegraphing on how some of the bash commands work bug me. Like why can it run some commands without asking me? I know why, I’ve seen the code, but that crap should be clearer in the UI. The first time it executed a bash command without asking me I was confused and somewhat livid because it defied my expectations. I actually read the crap it puts out because it couldn’t code its way out of a paper bag without supervision.

Not sure where you're looking, but you have companies like Minstral releasing Ministral 3 models for edge compute. Works fine on gaming PCs.

Edit: I doubt the "defaults" companies ship are going to matter much. We see what's happening with Microsoft and I've not met anyone that's happy with how LLMs are being shoveled into every digital product possible. Seriously, Microsoft doesn't need Copilot in their OS, they need to fix their stupid start menu decisions first.

You're wrong. The house literally will come up to someone that is winning a lot and tell them to leave. "You're too good, we have to ask you to leave."

This is pretty naive. What happens when you develop and extend such a system in a way that it can track who you interact with? What about social credit scores? You might go out to a social event with a very distinguished social credit score of 820 and get knocked down to 69 just because you were in proximity to Bob and Alice, who happen to be on some blacklists for their work in cryptography.

What you're staring at is the gateway tech that brings in a dystopian society. At first stuff like this is fairly benign, but slowly over time it ramps up into truly awful outcomes.

Nothing against you personally, just so you know. But I have to point out that anyone caring about the reason for the short coming of flock on stuff like this are just crafting soft reasons they can use to justify things later. Being up front here I care not for their reason because the entire business model is frankly disgusting and an affront to a functioning society. This is the type of tech that evolves into social credit scores and precog crime units, stoping crime before it happens.

At the end of the day your rationalization only affords comfort to those that have a vested interest in this stuff being successful and it needs to be clear to those people driving this that they’re not doing something popular or even good.

> You should probably read the comment you're replying to before replying

Congrats you spotted the thing we agreed on between comments. If you fail to see the agreement through parity of the part that was echoed, idk what to tell you. Education system is failing everyone in it these days.

You miss the point. This is a law enforcement tool. The average American doesn’t want a surveillance state and that’s literally what’s happening. The legal aspect of it is not in question here.

Just because something is legal doesn’t make it right. Anyone deploying or involved with this technology should be embarrassed and ashamed of themselves.