14 chunks · 12 episodes
...licated to be operated by anyone who wasn't a specialist. Now the LLMs have the background knowledge to execute problems that require specialist knowledge[b]. The situated knowledge, curiosity, and judgment is what is most important for the human to ...
...eone else's pitch, you need to be able to fill in the blanks. The more relevant background knowledge you have the more likely you have an approximate idea that fits. The more that your default understanding doesn't match what they're trying to tell y...
...on goes over most people's heads, only for the right specialists with the right background knowledge does it land.
...nto English it's unusable. There are a whole bunch of expectations and presumed background knowledge about techniques that are not explicit. All content presumes some background knowledge from the context it originates in and is expected to be consum...
...hen tag this doc into Claude conversations easily and give it extremely nuanced background knowledge when I'm trying to brainstorm on a problem. When someone wants to know what I'm working on, instead of sending a one-size-fits-none fossilized docume...
...s of last weeks' reflections. Often it's hard and tiresome to tell them all the background knowledge you need them to know. LLMs are in a dark room; each conversation is a fresh start where you have to start from scratch giving it background knowledg...
The power of LLMs comes from humans. Both the background knowledge that makes them smart is from culture[afy]. But also the thing that makes their output good is the quality of the steering the user is doing via prom...
...han a mouse?[akd] You can think of summarization as a process to factor out the background knowledge a reasonable listener would take for granted, leaving only the "diff" of interesting meaning. That implies that the more the LLM understands about th...
... the output. RAG can't give huge context to a model that doesn't have the right background knowledge, but it can be updated quickly and can enable precision in details. Everyone talks about these things like they're the same, but they're wildly diffe...
...terns like few-show learning. Fastest: Context engineering Precisely what extra background knowledge you pass it to help it answer this question. People talk about fine-tuning and making new foundation models, but the last two layers are remarkably e...
Training gives LLMs background knowledge. Context gives them working knowledge. Many people are worried about LLMs using their data in training, but there the leverage is way lower for the m...
...ent folksonomy that can grow itself by starting from a good-enough crystallized background knowledge. Now, instead of hyper-precise clockwork gears, you have rough clay that you can smoosh into place. More organic than mechanical.
...ider them more deeply. To people who aren't curious or don't have the necessary background knowledge, the elegant distillation will look perplexing, opaque, unrelated to the question they asked. But when you have the ability and time to make them blo...
...ter the narrative. In some cases your receivers will already have disconfirming background knowledge or priors "Wait a second, aren't you ignoring..." In that case, the narrative might not ever be adopted in the first place.