README.md

metadata

datasets:
  - stanfordnlp/sst2
language:
  - en
metrics:
  - accuracy: 0.91789

The model outputs 1 & 0. Where, 1 is "Positive" & 2 in "Negative".