@georgejrjrjr 2025-03-01 ♥0 ↻5 original ↗
lol thanks aidan.

would y'all please consider making the base model available?

GPT-3 and code-davinci-002 were awesome but tragically, abruptly yanked. G4B was insanely dope but few got to experience it, even second hand (NDAs prohibited posting outputs publicly).

I'd wager making a G4.5B experience available would be a 10x enthusiasm multiplier for high taste users, and diffuse some of the accumulated RLHFrustration.

And the theoretical safety implications hit different in a world where Grok DeepSearch and R1-Zero will make you nerve gas step-by-step.

author:georgejrjrjr kind:tweet model:code-davinci-002 model:deepseek-r1-zero model:gpt-3 model:gpt-4-base on:gpt-4-base year:2025

cited on: gpt-4-base

Reproduced against link rot, credited and linked to its original. Yours and you’d rather it weren’t here? Open an issue.