← Back to context

Comment by avarun

2 days ago

GPT-6 is supposed to be using a much larger base model that just finished pretraining so the "dump another chunk of all written language" approach is still going strong.

Modern pretraining also consists of expensive human-led specialized task creation and grading loops. Synthetic generation and distillation from previous models is another input for training. I wonder how much new text contributes beyond keeping knowledge up-to-date.

Where did they get a second copy of all books, written/audio/visual media, and the web?

Couch cushions?

The measure is to see if the results scale, not just the rumored attempts at building such a model, as o3 taught us.

Where are they getting new data from at this point? Didn't they already read the entire internet?

  • A lot of data these days is synthetically generated. As an example, to make a model good at understanding assembly you simply need to round-trip code through a compiler and disassembler, then train against the source of truth and the assembly. You can generate arbitrary algebra expressions and have it solve it.

    A lot of pretraining is also choosing the right type of data, you don't want to just have it ingest garbage (although I read that some amount of garbage actually helps the model be more robust). Pretraining crystallizes a lot of the inductive biases that post-training builds on, so by crafting the right data mixture you can make it easier for it to start off with a good foundation. There is also a lot of focus on mid-training these days, which I understand is basically either the name for the synthetic data stage, or the SFT phase before all the RL