Oddbean new post about | logout
▲ ▼
iefan 🕊️ | 1 months ago (raw) | export | reply | flag +6 
 A prompt that helps Claude 3.5 Sonnet beat OpenAI's o1 model in reasoning! https://image.nostr.build/88bbf240175554b202f853ba6228453ab8c598494175634fe5ed247c4b927288.jpg
▲ ▼
iefan 🕊️ | 1 months ago (raw) | root | parent | reply | flag 
 Testing it with my upcoming personality-based assistant, using Flash-8b, which is currently one of the fastest models. The results are phenomenal.
https://i.imgur.com/9bBCnd5.jpeg
nostr:nevent1qqsq72y32j707ugtgu26h77marmzrsu0t6dtfpw500q6j49fp3hzckgpzamhxue69uhhyetvv9ujuvrcvd5xzapwvdhk6tczyrr0wpmlz6va2r8e92t990ltl7kqtlrgg2u7uwgs38v4nw9dt4y06qcyqqqqqqgj7csyp
▲ ▼
Gzuuus | 1 months ago (raw) | root | parent | reply | flag 
 Interesting 👀 did you use it as a system prompt?
▲ ▼
iefan 🕊️ | 1 months ago (raw) | root | parent | reply | flag 
 Working on something
nostr:nevent1qqs0mr006vwv866frr6mzheqmmdhlflyv6yptmlvfz249esuzj87fhgpzamhxue69uhhyetvv9ujuvrcvd5xzapwvdhk6tczyrr0wpmlz6va2r8e92t990ltl7kqtlrgg2u7uwgs38v4nw9dt4y06qcyqqqqqqgyj9rt3
▲ ▼
Gzuuus | 1 months ago (raw) | root | parent | reply | flag 
 Awesome! I will test asa is accesible 👌
▲ ▼
Karnage | 1 months ago (raw) | root | parent | reply | flag 
 What tool is this for?