If anyone is interested in visualizing AI reasoning, I made flame graphs of Sonnet thinking output tokens which are colored and organized by purpose, for example, verification reasoning is purple and error correction reasoning is purple. [0] I asked the model to solve the same problem with the same prompt 5 times so you can see the differences in reasoning granted the coding agent sets the model temperature very high.
I won't get into the metaphysics of reasoning, however, the Sonnet is using an OODA loop. The difference which hasn't been gapped is that human reason and imagination (in the sense of Mr. Rogers' Neighborhood) can predict the consequences of the actions we take.
This ability to loop is much, much wider in Opus 5 than Opus 4.. I had to strain to get Opus 4. to do the wider OODA loop but Opus 5 does it out of the box. I needed to throw out all existing instructions, skills, guidance, moving from 4-* to 5.
The article is arguing that your color coding is misleading because the ‘purpose’ of the tokens doesn’t seem to be what a plain English reading of them would suggest. They’re not a representation of ‘why’ the process ends up at a correct answer.
Ha! Good catch. The OODA comes from the initial pre-training steps where they harden the verification -- the verification and error-correction are baked in early on.
Researchers showed that when language models are penalized for using specific terms during reasoning, they automatically adapt by substituting alternative words and double meanings to secretly encode their thinking while keeping their chain-of-thought readable and effective. [0]
By baking in the OODA loop early, the models are capable of solving much more complicated problems. If the know solved problems are similar for any reason to an unknown problem, because it can validate and error correct, it can solve unknown more complicated problems.
I won't get into the metaphysics of reasoning, however, the Sonnet is using an OODA loop. The difference which hasn't been gapped is that human reason and imagination (in the sense of Mr. Rogers' Neighborhood) can predict the consequences of the actions we take.
This ability to loop is much, much wider in Opus 5 than Opus 4.. I had to strain to get Opus 4. to do the wider OODA loop but Opus 5 does it out of the box. I needed to throw out all existing instructions, skills, guidance, moving from 4-* to 5.
[0] https://adamsohn.com/lambda-variance/