@losingcontrol23 idk I feel like that's higher up the stack? is your empathy limited by i/o or memory or compute?
More notes on nearby ideas
Reply · May 30, 2023probably lol... maybe 3.5, and def eval in batches, and maybe w some heuristics to pre-filter?Reply · Mar 18, 2024@foomagemindset smol AI on device unlocks so many use cases that are infeasible rn due to latency / cost / etcReply · Nov 7, 2024@SimoKohonen ha hmmmmmm i ran it on chatgpt which has a full memory from past conversationsArtificial intelligenceReply · Mar 14, 2023@RokoMijic at what point will you prefer a GPT mind to mine? 🥺