PinnedMike YounginGeek CultureMy Accidental Path to 1,000 Dev.to FollowersMy path to almost-accidental popularity on Dev.toJun 25, 20231Jun 25, 20231
Mike YoungCan you trick LLMs into giving dangerous info by phrasing questions in past tense?Doctors hate him! Get 8,000% more “dangerous” answers with this one weird trick!Jul 20Jul 20
Mike Young10–30% of today’s research is written by AIChatGPT is used for tons of science writing, especially in certain fields and countriesJul 12Jul 12
Mike YoungWhat if we recaption billions of images?Less noise, 5x caption length, and 9x LongCLIP scores.Jun 23Jun 23
Mike YoungActually, telling ChatGPT to “think step-by-step” doesn’t help muchChain-of-thought prompting isn’t really that usefulJun 4Jun 4
Mike YounginArtificial Intelligence in Plain EnglishMachines curate better data for themselves than a human wouldLet AI pick its own training dataMay 28May 28
Mike YoungPoking parts of an AI’s brain to make it less annoyingWe now have a map to where concepts live in the minds of LLMsMay 22May 22
Mike YoungMeet Chameleon. Is it Meta’s answer to GPT-4o?In the wake of OpenAI’s recent announcement of GPT-4o, a new model that processes and generates text, audio, and images in real-time, it’s…May 17May 17