Comment by bumby

To be a fair comparison wouldn’t you have to include all the energy used to train “your” brain through generations of evolutionary training? Your latest model is like taking an already trained BERT model and adding a few tweaks