Hacker News new | ask | show | jobs
by whstl 25 days ago
If compaction is throwing away crucial prompting instructions even when it's at a 1% of maximum token usage (like my example), then it's a software bug, not an LLM artifact.
1 comments
tacone 25 days ago
Doesn't compaction invalidate token caching, btw?
link
whstl 25 days ago
I don't see how this has anything to do with my message, sorry.
link