Video yükleniyor...

Video Yüklenemedi

Ana Sayfaya Dön

How Large Language Models Store Facts

51,698 görüntüleme • 1 yıl önce •via X (Twitter)

8 Yorum

Ronen Lahat profil fotoğrafı
Ronen Lahat1 yıl önce

@3blue1brown This is what the internet is for, watching two 3blue1brown llm videos simultaneously

Mohan LM profil fotoğrafı
Mohan LM1 yıl önce

@3blue1brown thx where iz the rest of the video, pls.

Girish profil fotoğrafı
Girish1 yıl önce

@3blue1brown What a coincidence, i was just discussing this very topic last night

lxos.xivs profil fotoğrafı
lxos.xivs1 yıl önce

@3blue1brown Honestly if you need these animation to understand these concepts. You don’t need to understand them

josejuan profil fotoğrafı
josejuan1 yıl önce

@3blue1brown It describes a config of weights, usually tied to a layer/embedding. The "facts" in an ANN are distributed across the network, with nonlinearity between layers being crucial for capturing relationships like "big fish eats little fish," which can't be stored in a single embedding.

Quentin Munch profil fotoğrafı
Quentin Munch1 yıl önce

@3blue1brown Store data* (not facts)

Dannie Sim profil fotoğrafı
Dannie Sim1 yıl önce

@3blue1brown nailed it

Light 𓆩𓆩♡𓆪𓆪 profil fotoğrafı
Light 𓆩𓆩♡𓆪𓆪1 yıl önce

@3blue1brown nearly perpendicular

Benzer Videolar