Pietro Lesci
pietrolesci
·
AI & ML interests
I like developing and applying causal methods to study the effect of training choices on models’ behaviour, including memorisation, shortcut learning, and tokenisation.
Recent Activity
upvoted
an
article
24 days ago
There is no such thing as a tokenizer-free lunch
updated
a model
about 1 month ago
pietrolesci/tokenisers
updated
a model
about 1 month ago
pietrolesci/tokenizers