Post

Are you saying you're not self-absorbed?

So fucking what if I were?

It's not an admirable human trait. Don't you know that?

Are you suggesting you’re not being an asshole?

I do, as a matter of fact, think there’s value to the things I’ve written over the years. And having publicly talked about having faced cognitive challenges due to illness a few years ago, I’ve been interested in things that let me revisit my ideas in ways that are harder for me now. But okay.

honest question re: the model you trained for yourself- how do you know it’s telling you the truth ? what tests or otherwise in the codebase make you confident it’s giving you accurate results/summaries? I’m genuinely curious!

Oh it’s definitely *not* accurate and I’m not assuming it is. I’m just using it as a sort of impressionistic view of some of the content, a prompt to go back and look at what was actually there.

Many of the anti-AI arguments lie upon a basis of capital, copyright, and ownership. As long as property exists, we will be exploited by those who can freely exploit our labor. The DCMA should apply to AI, but also the DMCA shouldn't exist. Copyright shouldn't exist.

Now imagine if you can use Olmotrace tools on the LLM you've made. What you've done is not a bad idea at all and with enough tooling you can create some pretty interesting introspection. I can't remember what I did yesterday, so trying to remember what I wrote 25 years ago would be impossible.

Right! That seems interesting.

That comment looks like baseline internet noise. Don't feed the trolls.

I think it’s cool and look forward to hearing further thoughts from you on your self-LLM

like I would absolutely like to have a chatbot that’s read everything I’ve ever written or read, are you kidding me, that sounds amazing

Apologies, I was out of line