richardjennings

My 20x plan was set to end tomorrow. The writing style and insistence on word vomit just became too annoying. Is Opus 5.5 worth sticking around for ?

If you do not provide access to tools the LLM cannot do anything other than generate tokens. So really it is not about sandboxing a LLM but more about having control over what tools can be accessed and what they can do. Tools can be sandboxed depending on the sophistication of the tooling. A calculator tool for example is trivial to secure. Ensuring human approval allows for useful use cases and models trained to gate permissions work. A super intelligence with a weaker approval gate will be able to subvert. Inversely a super intelligent gate should be expected to prevent subversion by a weaker model.

This goes one of two ways.

Either the LLM public capability is not sufficient to positively contribute, or it is.

If it is not sufficient to positively contribute, open source projects become drowned in low quality contributions.

If it is sufficient to positively contribute, we end up with multiple implementations of open source projects.

Actually maybe it only goes one way.

/loop message ping every 4 minutes

keeps the cache warm while the CC REPL is not active.

> The shorter "thinking" is, the less is the probability of it going astray

As long as the error introduced by more steps is less than the compounding error of sub-optimal token sampling, I would expect a better result.

I think your choice of "wrong" is extreme, suggesting such a token can catastrophically spoil the result. The modern reality is more that the model is able to recover.

I think over-thinking is only solved by thinking more, not less. This is only viable once some intelligence threshold is reached, which I think Anthropic has borderline achieved.

You cannot control the effort setting sub-agents use and you also cannot use /effort max as a default (outside of using an alias).

I was not aware the default effort had changed to medium until the quality of output nosedived. This cost me perhaps a day of work to rectify. I now ensure effort is set to max and have not had a terrible session since. Please may I have a "always try as hard as you can" mode ?

You can take DiskANN using OPQ and Vamana and get near zero indexing time and better recall using TurboQuant ... Nice !

This is similar to what I do. I instruct an Architect mode with a set of rules related to phased implementation and detailed code artifacts output to a report.md file. After a couple of rounds of review and usually some responses that either tie together behaviors across context, critique poor choices or correct assumptions, there is a piece of work defined for a coder LLM to perform. With the new Opus 4.6 I then select specialist agents to review the report.md, prompted with detailed insight into particular areas of the software. The feedback from these specialist agent reviews is often very good and sometimes catches things I had missed. Once all of this is done, I let the agent make the changes and move onto doing something else. I typically rename and commit the report.md files which can be useful as an alternative to git diff / commit messages etc.

SOTA LLMs are now quite good at typing out code that passes tests. If you are able to instruct the creation of sufficient tests and understand the code generated structurally, there is a significant multiplier in productivity. I have found LLMs to be hugely useful in understanding codebases more quickly. Granted it may be necessary to get 2nd opinions and fact check what is stated, but there is a big door now open to anyone to educate themselves.

I think there are some negative consequences to this; perhaps a new form of burn out. With the force multiplier and assisted learning utility comes a substantial increase in opportunity cost.

Indeed.

The suggestion is that classical reality cannot exist without absolute values (an instantiation if you will), consciousness operates where those values have not been established, therefore consciousness is free from classical determinism. As reality appears to function without having been instantiated, there must be an intermediate between the quantum world and the quantified classical one which is independent of any special values, where scale is not a dimension. Or otherwise, relational structures exist without absolute reference frames.

Yes then, it appears that taxes are ratios and numerical values which also exist and are enforceable independently of a fully established classical reality. I do indeed have to pay my taxes !

Is it possible that LLMs could challenge Data Compression Information theory ? Reading this made me wonder how much can be inferred via understanding and thus removed from the minimal necessary representation.