A calm explainer on Transformer (deep learning architecture) for absolute beginners

Hana Choi · 43 likes

That is a good question for people with more experience than me. The Toronto example makes it feel concrete. The boring solution is probably the reliable one here

Cozy Marauder · 39 likes

what would you change if you started over? I would test it for 5 days before deciding

Lotte Jansen · 12 likes

Helpful and not preachy which is rare. I noticed the same thing in Cape Town

Jack Taylor · 2 likes

could be regional too

Grace Parker · 9 likes

The absolute angle is what sold me. It reminds me of a very specific conversation near the kitchen

Aisha Jacobs · 9 likes

This sounds sensible, though I would need to test it myself. I learned that one the expensive way

Aoi Ito · 5 likes

Agree completely. The simple version is underrated. The boring solution is probably the reliable one here

Cozy Noodle · 4 likes

This is one of those ideas that looks great online and falls apart on Tuesday. This is where I would ask one more question before choosing

Midnight Fern · 4 likes

Great point, especially the part about deep. This would have saved me a lot of trial and error

Ngozi Bello · 2 likes

I think teh bad version of this advice causes alot of frustration. I would love to see a follow-up after 5 weeks

Mateo Cruz · 1 likes

This has cozy competence written all over it. The quiet Sunday version is even more interesting

Kabir Khan · 5 likes

noted, and thank you

Chloe Davis · 3 likes

The best answer in the thread, honestly

Valeria López · 3 likes

This comment has main-character footnote energy

Kabir Khan · 2 likes

I laughed, but you are not wrong

Noah Miller · 1 likes

I was going to disagree, but your second sentence got me

Amelia Hughes · 1 likes

maybe, but the tradeoff seems bigger than the post suggests. The 7-minute rule is underrated

Tola Ibrahim · 1 likes

This reminds me of something a friend in Istanbul mentioned. That is where the boring checklist helps

Avery Johnson · 1 likes

big fan of this approach. I keep a note about prompts for exactly this reason