Also 95% of those 30k lines changed are fully self-contained inside of the aarch64 directory and of the remaining changes it looks like the majority is just adding "aarch64" as another item into an existing list. There are a few core changes that to me look like they could be done in their own PRs, but also core maintainers get to decide if they want to apply bureaucracy to their own work.
I don't understand that methodology in the first place. Does Anthropic even have some kind of somewhat objective definition to measure and judge "memorization"? Is there any evidence that other LLMs are viable tool to determine that?
If we were perfectly rational, then yeah, more data should never lead to worse decisions. However, it's easy to fall into the trap where being data-driven makes you only work on those things that you know how to measure.
While xor eax, eax only uses 2 bytes. Since there are only 8 registers, meaning they can be encoded with 3 bits, you can pack two values into the <Registers> field (ModR/M).
Making mov eax, 0 only take two bytes would require significant changes of the ISA to allow immediate values in the ModR/M byte (or similar) but there would be little benefit since zeroing can already be done in 2 bytes and I doubt that other cases are even close to frequent enough for this to be any significant benefit overall. An actual improvement would be if there was a dedicated 1 Byte set-rax-to-0 instruction, but obviously that comes at a tradeoff where we have to encode another operation differently (probably with more bytes) again (and you can't zero anything else with it).
Some other architectures like PDP-11 and 680x0 had a dedicated "clear register" instruction.
It could have been added to x86, even as a group of single-byte opcodes with the register encoded in three bits (as with PUSH, POP, and INC/DEC outside of long mode). But the XOR idiom was already established on the 8080 by that point.
I would be surprised if modern CPUs didn't decode "xor eax, eax" into a set of micro-ops that simply moves from an externally invisible dedicated 0 register. These days the x86 ISA is more of an API contract than an actual representation of what the hardware internals do.
The predominance of these idioms as a way to zero out a register led Intel to add special xor r, r-detection and sub r, r-detection in the instruction decoding front-end and rename the destination to an internal zero register, bypassing the execution of the instruction entirely. You can imagine that the instruction, in some sense, “takes zero cycles to execute”.
We have not been shown any evidence that shows anything approaching this scale of people dead, we have not been told on what basis that specific number of deaths was determined and we have not been told why that supposed evidence can not be made public. 30 thousand killed in the span of two days would make this one of the bloodiest massacres in history. I don't see how that would not result in at least a single photo or satellite image of a street lined with dead bodies, of mass executions, of mass graves or simply just the logistics operation required to dispose of this many bodies. The only thing to back up this number is trusting the probably least trustworthy US administration in history.
During Covid, there were satellite pictures of mass graves. There is a certain amount of logistics with moving a lot of dead bodies, that is hard to hide.
> This also re-opens a lot of "party pooper" results in mathematics: impossibility of representing solutions to general quintic (fine print: if we restrict ourselves to arithmetic and roots/radicals).
Solving the quintic in terms of transcendental functions has already been done
reply