i realized i had terminal ai brainrot when i asked the llm to change all the model calls from oai to anthropic when i could have just done find and replace
We tested a pre-release version of o3 and found that it frequently fabricates actions it never took, and then elaborately justifies these actions when confronted.
We were surprised, so we dug deeper ๐๐งต(1/)
https://t.co/IdBboD7NsP
Introducing... DolphinGemma! ๐ฌ
๐ฌNext dolphin token predictor model (audio-to-audio)
๐คJust 400M parameters - can run directly in phones!
๐๏ธFin-tuning possible for different species
๐To be released for the next dolphin season
https://t.co/qGoHGVegGK