HN
Hacker News
Top
New
Best
Ask
Show
Jobs
Comment by unleaded | Hacker News Reader
Full thread
unleaded
·
ITT nobody remembers gpt2 anymore and that makes me sad
View on HN
GaggiX
·
This model was trained on 6T tokens and has 256k embeddings, quite different than a gpt2 model comparable in size.
Reply on news.ycombinator.com