Ranveer Shah
@Ranveer3sI like building stuff and working with tech | ⚽ | ♟️
How it must feel to be a latent vector flowing through transformer attention layers
We beat Barca in the Europa League 2 seasons ago. Fast forwarded to 2024, Barca are in the UCL, beat Bayern 4-1 in the UCL, just thrashed Real Madrid 4-0 at the Bernabeu, and sitting top of the league with 27 GD. Meanwhile, Man Utd couldn't beat Twente at home in the Europa…
This is great
Always a good read
If you know who the person on the left is, but not the person on the right, then you have been utterly failed by the media that you consume
For almost any big project, my two most common unblocking questions are: 1. How would I do this, as intelligently as possible, with years of time and unlimited money? 2. How would I do this, as stupidly as possible, in very little time and with very little money?
Pre-training Small Base LMs with Fewer Tokens - Inherit a few layers from the larger LM and then train on a very small subset (0.1%) of the pretraining data - Can effectively match their bigger counterparts repo: github.com/sanyalsunny111… abs: arxiv.org/abs/2404.08634
An interesting AI-based repository maintainer I came across github.com/apps/dev-ape-a…
United States Trends
- 1. #GivingTuesday 19,1 B posts
- 2. Good Tuesday 31 B posts
- 3. #tuesdayvibe 3.246 posts
- 4. #alieninvasion N/A
- 5. Elvis 17,4 B posts
- 6. #TuesdayFeeling 1.031 posts
- 7. #3Dic 2.355 posts
- 8. Delaware 73,3 B posts
- 9. Jameis 63,2 B posts
- 10. Daniel Penny 84,5 B posts
- 11. Manfred 4.729 posts
- 12. Starfire 1.875 posts
- 13. Lemmy 1.253 posts
- 14. Watson 21,2 B posts
- 15. Vermont 16,4 B posts
- 16. Golden At-Bat 1.074 posts
- 17. Roddy 4.541 posts
- 18. Judges 72,4 B posts
- 19. Jeudy 31,9 B posts
- 20. Bovaer 84,3 B posts
Something went wrong.
Something went wrong.