In a recent post, Zvi described what he calls "The Most Forbidden Technique":
An AI produces a final output [X] via some method [M]. You can analyze [M] using technique [T], to learn what the AI is up to. You could train on that. Never do that.
You train on [X]. Only [X]. Never [M], never [T].
The article specifically discusses this in relation to reasoning models and Chain of Thought (CoT): if we train a model not to admit to lying in its CoT, it might still lie in the CoT and just not tell us.
This seems like a good idea, but something I read recently makes me wonder if we can really avoid this.
When you're subject to capital gains taxation, the government shares in some of the upside, but when you have capital losses, the government shares in the downside too. Because of this, the actual risk (and reward) of any given portfolio is lower than it seems. To counteract this, you should consider shifting your allocation toward riskier assets.
7 years ago (!) I wrote a post comparing ZIP and tar, plus gz or xz, and concluded that ZIP is ideal if you need to quickly access individual files in the compressed archive, and tar + compression (like tar + xz) is ideal if you need maximum compression. Since then, I discovered pixz, which seems to provide the best of both worlds: Maximum compression with indexing for quick seeking.
Do you have an insatiable hunger for potatoes? Do you have a fully-sated hunger for complicated recipes and having to cook all the time? This version of Pommes Anna, simplified and altered to an extent that will make French people cry, might be the recipe for you!
Psyllium husk is a non-fermenting (no gas or bloating) soluble dietary fiber that improves both constipation and diarrhea (such as with IBS), normalizes blood sugar, reduces LDL ("bad") cholesterol, and can help with weight loss. Each type of dietary fiber has different effects, and a "high fiber" diet in general won't necessarily provide the same benefits, especially for conditions like Irritable Bowel Syndrome.
