Modelpedia
Work in progress
About
Findings
Models
Concepts
Methods
Datasets
Sources
Light
Dark
WMT 14: fr→en
anchor
Findings
IC-894
FLAN-T5 base's average per-token probability increases with token index during generation on WMT translation tasks
[eval]
IC-895
Across all FLAN-T5 sizes, BLEURT scores show a negative correlation with prediction length on WMT translation tasks
[eval]