Viewing profile — Catloafdev

How is this relevant?

Trusting snake oil is a wild choice.

I mean what are these devs doing that they are day-1 updating npm packages still? How many instances of this are required before npm package maintainers learn?

Your comment was fine, I didn't think you were implying that at all. I was just trying to clarify that they aren't choosing to do that for arbitrary reasons.

Neat! But, what do you do with a 0.5tk/s LLM? Have you tried running it via llamacpp or other software that supports naive SSD offloading to compare speeds?

The problem is the 'options to be intentionally non-destructive' come with legal liabilities that don't exist if they destroy the original copy.

The scanning process they use is _intentionally_ destructive, not accidentally destructive. It's not because they hate books or something. It's literally to stay in the law due to …

If you bothered to read the article, you'd understand why this is, in fact, a problem.

Weird that you comment on rage bait and end your comment with the idea that this group of people are all a bunch of worthless drug users.

If any editors are reading this thread, the weights are on HF now, but the 'open source' question still implies it's proprietary/API-only.

> Agent code can only be read and understood by agents. This just fundamentally isn't true and if this is your perspective then you're using LLMs wrong.

That's LOC in a single file, that's a bit different from total LOC as a metric. Total LOC is a garbage metric. Things like reducing line count in specific files or components is a …

[dead]

There quite literally are internals that are called attention heads in modern LLMs. Saying 'there are no attention heads' is objectively false, which definition you follow is irrel…

Ya I'd be interested to see a comparison of using llamacpp with ssd offloading to compare real speeds.

> There are no "attention heads" Yes there quite literally is internally in an LLM.

How is this different from Obsidian? Why would someone switch? The page is a bit lacking on detail.

It's absolutely reasonable to have safeguards on sufficiently dangerous models being released - if you disagree, can you explain your perspective? I think it's wildly irresponsible…

> I would say any AI generated output... would have been extremely easy to detect > I write everything using AI and nobody has detected it Might need to rethink this one, Googling …

Why are you assuming they aren't? The people who have those books are choosing to sell them for a price. What isn't fair?

I'm not sure I'd use the term 'brittle' to describe snake oil.

Reassuring to see this backed by major players. This is the correct stance, hopefully this is the stance that prevails.

The same authority he used to enact tariffs and a war.

Well it seems to be hug-of-death'd - how are you loading 41k points of data?