Outer Web | Explore

/explore

Click through on any links that interest you or select the planets on the right to continue exploring the Outer Web.

You are here		www.alignmentforum.org New Scaling Laws for Large Language Models - AI Alignment Forum
\|	\|	deepmind.google An empirical analysis of compute-optimal large language model training - Google DeepMind	0.9 parsecs away Travel
\|	\|	We ask the question: "What is the optimal model size and number of training tokens for a given compute budget?" To answer this question, we train models of various sizes and with various numbers...	0.9 parsecs away Travel
\|	\|	www.lesswrong.com The Agency Overhang - LessWrong	2.0 parsecs away Travel
\|	\|	As language models become increasingly capable of performingcognitive tasks that humans can solve quickly, it appears that a "cognitive capability :...	2.0 parsecs away Travel
\|	\|	blog.moonglow.ai Three Kuhnian Revolutions in ML Training	1.9 parsecs away Travel
\|	\|	Parameters and data. These are the two ingredients of training ML models. The total amount of computation ("compute") you need to do to train a model is proportional to the number of parameters multiplied by the amount of data (measured in "tokens"). Four years ago, it was well-known that if	1.9 parsecs away Travel
\|	\|	conorneill.com The hidden danger of "I am fine now" - Moving People to Action	5.2 parsecs away Travel
\|		Often when I ask a leader if they have a coach, or are seeking training, the answer is something along the lines of "I don't need a coach... I am fine now". This is a dangerous approach. The farmer doesn't wait for winter to repair the roof. The farmer doesn't wait for winter to sew	5.2 parsecs away Travel