Swayson
9.1K posts

Swayson
@Swayson
Thinker. Tinkerer. The Curious. Fascinated by Cognitive Science, AI, Philosophy and Nature.


We really are living through the singularity aren't we.







My take on why LLMs still suck at forecasting is that in some cases they are still over-reliant on pre-training priors and that they are getting fried by post-training / RL and safety, which makes them non-committal and hedgy. But I also think it's just not a skill that they are picking up in any math or coding envs. In the GPT-4 there was a slide on the calibration of base-model vs the final post-trained model. The base model was almost perfectly calibrated, the post-trained model wasn't. (this is of course weak'ish evidence because we don't know whether this is still true with current models and how its calibration translates to forecasting performance)









this must have been so satisfying










