Certainly, you’re right that reading it as “konnichi” is very rare. It’s rare enough that I only sometimes misread it that way when it appears in the news or similar contexts. So I think that was a good decision.
By the way, I tried testing it further while thinking back to the kinds of tests I had when I was in school. The accuracy is still excellent. My guess is that “一日” and “分別” are being handled in a similar way to “今日.” “分別” is very rare, but I don’t think “一日” is all that uncommon.
I’m Japanese. I was surprised that it was able to answer correctly even when I entered commonly seen difficult-to-read place names. However, there seem to be cases where it may incorrectly read “今日” when it should be read as “こんにち.”
Example: 今日の日本社会では、少子高齢化が大きな課題となっている。
Also, it’s disappointing that Japanese does not appear even when I select it.
Please let me know if there’s anything I can do to help.
I recently consolidated the fronted, backend, and infrastructure repositories using submodules, and Claude Code can handle it very well without external tools.
The newer modules have a long context window, so they can understand even huge pretty well.
Nice concept!
I’m concerned about that LLM might discover `faz.yaml` and directly access the databases.
Wouldn’t it be more deterministic and safer to wrap the database itself and use a safety-pipeline-enabled DB instead?
Where is `verify` and `secureWipe` func?
I just found `unsecureWipe` func as `secureWipe` func in argon2-swift.
Did you just forget to remove an internal state?
By the way, I tried testing it further while thinking back to the kinds of tests I had when I was in school. The accuracy is still excellent. My guess is that “一日” and “分別” are being handled in a similar way to “今日.” “分別” is very rare, but I don’t think “一日” is all that uncommon.