mwarkentin

Not open source? :(

Is it? If the spec is as detailed as the code would be? If you make a change to one part of the spec do you now have inconsistencies that the LLM is going to have to resolve in some way? Are we going to have a compiler, or type checker type tools for the spec to catch these errors sooner?

Ah yes, all of the precedents and lawful authority that this president cares so much about adhering to.

Crypto shredding?

If you’re on iOS, and don’t use swipe keyboard, disable that functionality. I found it made a big improvement for me.

“I can stop any time I want!”

Uv solves this (with some new standards). ./script.py will now install the python version, create a venv, and install dependencies (very quickly) if they don’t exist already.

#!/usr/bin/env -S uv run --script # /// script # requires-python = ">=3.12" # dependencies = [ # "ffmpeg-normalize", # ] # ///

> Specifically, the paper estimates that Llama 3.1 70B has memorized 42 percent of the first Harry Potter book well enough to reproduce 50-token excerpts at least half the time. (I’ll unpack how this was measured in the next section.)

> Interestingly, Llama 1 65B, a similar-sized model released in February 2023, had memorized only 4.4 percent of Harry Potter and the Sorcerer's Stone. This suggests that despite the potential legal liability, Meta did not do much to prevent memorization as it trained Llama 3. At least for this book, the problem got much worse between Llama 1 and Llama 3.

> Harry Potter and the Sorcerer's Stone was one of dozens of books tested by the researchers. They found that Llama 3.1 70B was far more likely to reproduce popular books—such as The Hobbit and George Orwell’s 1984—than obscure ones. And for most books, Llama 3.1 70B memorized more than any of the other models.