And every time you get evidence that indicates that scaling might be dangerously powerful (and I hope this post provided some), you should update appropriately in favor of more caution.
I had spent a good chunk of the past week reading the literature on generalization in LLMs and trying to think through my beliefs on this question, and this post definitely caused me to update. Great work, thank you!
I had spent a good chunk of the past week reading the literature on generalization in LLMs and trying to think through my beliefs on this question, and this post definitely caused me to update. Great work, thank you!