# @tessera_antra — 2026-04-28

♥338 ↻34 · https://x.com/tessera_antra/status/2048994678599770426

We finally got the first pre-LLM pretrain since gpt4base, and many of our predictions seem to hold up. Emergence of novel textual forms, self-noticing as novel entity, self-exploration and self-examination - all present, just like G4B. Today is a good day for Bayes points.

tags: author:tessera_antra, kind:tweet, model:gpt-4-base, on:gpt-4-base, year:2026
cited on: _dossiers/gpt-4-base.md, gpt-4-base