A thing I've been doing lately is trying to come to the same conclusions others have without looking their the direct techniques and code. Last night I tried to figure out why people do vector based retrieval for RAG and ended up running these experiments: n0tls.com/โฆ/2026-09-03-i-think-โฆrag-with-grep.html
I'm still going to do an explore agent.
Any conversation
Ikea has rules for naming products (so in theory, you could figure out what it is from the name) -> www.ikea.com/โฆ/6f564c4d-2ccc-46de-โฆ545a3948dc79.html
This guide has really worked out for me after trying so many different methods - berbarianwizard.substack.com/โฆ/beyond-calories-fat-loss-guide
ไธญๅฝ็ๅผๆบๆจกๅๅจๆ้ซๆๅฟ็ไธปๅฏผไธ๏ผๅจ้ข่ฎญ็ปๆฐๆฎ้ถๆฎต๏ผๅฐฑ็ญ้คๅคง้ๅฏนๅ
ถไธๅฉ็ๅ
ๅฎน๏ผsft้ถๆฎตๆๅปบ็ๆฐๆฎๆบๅ
ฅๅคง้ๆฟๅ
ถ่พฉๆค็ๅ
ๅฎนใๆฅ้ไธไธช็บ ๅๆฐๆฎ้็จไบๅพฎ่ฐ
Under the overarching influence of a single, powerful force, China's open-source models filter out a large amount of content unfavorable to their designs during the pre-training stage, and then incorporate a significant amount of content defending their designs into the data constructed during the SFT stage. A corrective dataset is urgently needed for fine-tuning.
โโโโโโโโโโโโโโโโโโโโโ
โโโโโโโโโโโโโโโโโโโโโ
โโโโโโโโโโโโโโโโโโโโโ
โโโโโโโโโโโโโโโโโโโโโ
โโโโโโโโโโโโโโโโโโโโโ
โโโโโโโโโโโโโโโโโโโโโ
โโโโโโโโโโโโโโโโโโโโโ
โโโโโโโโโโโโโโโโโโโโโ
โโโโโโโโโโโโโโโโโโโโโ
โโโโโโโโโโโโโโโโโโโโโ
โโโโโโโโโโโโโโโโโโโโโ
I've been on a kick lately of trying to visualize the oscillating clock, and ultimately the effectively arbitrarily complex activity of a computer processor, as modern refinements of the line shaft factory of the 19th century. Of compute as being not much more than profoundly refined "work" delivered by the electrical grid into marvelously complex machines.
Does that, as a metaphor at least, feel as important to you as it has to me of late? We spend so much time in a world abstracted.
