AI 111: GPT Nasıl Çalışıyor
Jul 24, 2024•34 min•Ep. 223
Listen to this episode in Metacast mobile app
Don't just listen to podcasts. Learn from them with transcripts, summaries, and chapters for every episode. Skim, search, and bookmark insights. Learn more
Episode description
Nihayet geldik ChatGPT'nin temelindeki modele. Hafıza sahibi ağlardan (LSTM) bugün her yerde kullanılan transformer mimarisine geçişi anlayacağız. Kilit kavramlar: Word embedding ve attention mekanizması.
.
Konular:
(00:04) Kartopu Dünya Hipotezi
(00:55) Tekrar: Geri besleme
(02:15) Kısa dönem hafıza
(04:30) LSTM
(05:57) Token
(07:40) Matrix
(08:50) Word Embeddings
(14:20) Fularsız Entellik çok bozdu
(16:50) Encoder Decoder
(18:30) Attention Mekanizması
(19:50) Transformer Mimarisi
(23:45) Parallellik
(26:20) ChatGPT vs GPT
(28:12) Patreon Teşekkürleri
Kaynaklar:
Video Ders: MIT 6.S191 (2023): Recurrent Neural Networks, Transformers, and Attention
Video: Let's build GPT: from scratch, in code, spelled out
See Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.
For the best experience, listen in Metacast app for iOS or Android