For me the Othello GPT paper was an inflection point in buying that very rich representations and latent programs emerge from next token prediction pretraining arxiv.org/pdf/2210.13382
For me the Othello GPT paper was an inflection point in buying that very rich representations and latent programs emerge from next token prediction pretraining arxiv.org/pdf/2210.13382
52
13
211
33K
135