This story was originally published on HackerNoon at:
https://hackernoon.com/understanding-scaling-law-through-data-science-lenses.
Despite the immense promise of LMs, initial endeavors to apply pre-trained LMs to downstream tasks have encountered significant challenges.
Check more stories related to data-science at:
https://hackernoon.com/c/data-science.
You can also check exclusive content about
#data-science,
#scaling-law,
#artificial-intelligence,
#tokenizing-raw-text,
#discrete-tokens,
#embedding-vector,
#token-embeddings,
#power-laws, and more.
This story was written by:
@tianchengxu. Learn more about this writer by checking
@tianchengxu's about page,
and for more stories, please visit
hackernoon.com.
Despite the immense promise of LMs as task-neutral foundation models, initial endeavors to apply pre-trained LMs to downstream tasks encountered significant challenges.