Skip to content
Better HN
Top
Best
Ask
Show
New
Jobs
Search
⌘K
0 points
madduci
6mo ago
0 comments
Save
Share
The first users of this dataset will be Big Tech corps. Meta, Alphabet, OpenAI, Microsoft, Apple will all be happy to use this dataset for training their LLMs.
For them, 300TB is just cheap
0 comments
1 comments · 1 top-level
top
newest
oldest
ipsum2
6mo ago
They already have this data. See jukebox from OpenAI, released before chatgpt.
j
/
k
navigate · click thread line to collapse
