I find it fascinating how similar the answers from different models with similar parameter counts are. Seems like an indication that they’re all modelling simil

I find it fascinating how similar the answers from different models with similar parameter counts are. Seems like an indication that they’re all modelling similarly datasets to a similar degree of fidelity.