Great content, as always! Just a quick off-topic question: My OKX wallet holds some USDT, and I have the seed phrase. (air carpet target dish off jeans toilet sweet piano spoil fruit essay). How should I go about transferring them to Binance?
Pretty neat I'll just have a question : You first explained that you choose word based encoding over character based because it needs a sequence model to make the difference between 2 similar sets of codes (2:13). But in the end (8:37), you also explain that next step will be about sequences of tokens (which make sense, because there's no meaning without order in a sentence). Why highlighting this difference while we could as well have dealt with character based encoding from the start ? This point is more about the didactic than the subject itself ^^
At the end we're highlighting that we'll show how to turn word tokens into sequences of word tokens to represent the sentences. So, 'I love my dog' instead of being a bag of words would be tokens in that order.
For the *how* -- it's best to look at the source code. For the *why*, if you're not going to keep everything, you, logically, want to keep the most common words in it.