1:14:50 For Cart Pole, I just took a look at their Jupyter Notebook, and realise they did it by predicting action at each timestep, so this part is similar to Decision Transformer. However, for embeddings, they just take the entire text of the sequence and embed it. This is suboptimal, as some numbers may be chunked together and some may not. It would be better if the embeddings are done per number instead of leaving it to the tiktoken tokeniser!
1:14:50 For Cart Pole, I just took a look at their Jupyter Notebook, and realise they did it by predicting action at each timestep, so this part is similar to Decision Transformer.
However, for embeddings, they just take the entire text of the sequence and embed it. This is suboptimal, as some numbers may be chunked together and some may not. It would be better if the embeddings are done per number instead of leaving it to the tiktoken tokeniser!
Great walk through.
Part 2 here: th-cam.com/video/rZ6hgFEe5nY/w-d-xo.html