bbunzeck/grapheme-llama
Text Generation • 15.1M • Updated • 3 • 1
Small Language Models Also Work With Small Vocabularies
Models and evaluation data for our 2025 COLING paper (https://aclanthology.org/2025.coling-main.404/).
-
-
bbunzeck/grapheme-llama-no-whitespace
Text Generation • 15.1M • Updated • 3 -
bbunzeck/phoneme-llama
Text Generation • 15M • Updated • 6 -
bbunzeck/phoneme-llama-no-whitespace
Text Generation • 15M • Updated • 4 -
bbunzeck/phoneme-babylm-10M
Viewer • Updated • 3.92M • 10 -
bbunzeck/phoneme-babylm-100M
Viewer • Updated • 15.8M • 39 -
bbunzeck/phoneme-blimp
Viewer • Updated • 59.9k • 102 -
bbunzeck/rhyme-sentences
Viewer • Updated • 400 • 20 -
bbunzeck/wug-words
Viewer • Updated • 1k • 9 -
Small Language Models Also Work With Small Vocabularies: Probing the Linguistic Abilities of Grapheme- and Phoneme-Based Baby Llamas
Paper • 2410.01487 • Published