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.
Organizations
models 27
pietrolesci/tokenisers
Updated
pietrolesci/tokenizers
Updated
pietrolesci/small_bpe128k
Updated
pietrolesci/small_langspec128k
Updated
pietrolesci/small_unigramlm128k
Updated
pietrolesci/small_tokmix128k
Updated
pietrolesci/small_multigram128k
Updated
pietrolesci/me100M_finewebedu-20B_bpe32000minipile
Updated
pietrolesci/me100M-tied_finewebedu-20B_bpe32000minipile
Updated
pietrolesci/me850M_minipile_bpe32000minipile
Updated
datasets 56
pietrolesci/unimixlm
Viewer
• Updated
• 81.9M • 278
pietrolesci/me-minipile-evals
Viewer
• Updated
• 1.22M • 27
pietrolesci/pile-deduped
Viewer
• Updated
• 748M • 8
pietrolesci/pythia-deduped-memorisation-profiles
Viewer
• Updated
• 2.13M • 37
pietrolesci/pile-validation
Viewer
• Updated
• 429k • 83
pietrolesci/pile-deduped-subset
Viewer
• Updated
• 16.3k • 9
pietrolesci/pythia-deduped-stats
Viewer
• Updated
• 16.3M • 34
pietrolesci/pythia-deduped-stats-raw
Viewer
• Updated
• 14.9M • 33.7k
pietrolesci/agnews
Viewer
• Updated
• 510k • 82
pietrolesci/amazoncat-13k
Viewer
• Updated
• 5.99M • 1.09k • 1