INDEX
Explanations
a more comprehensive overview
New Auto-Interp
Negative Logits
précis
0.90
recipe
0.89
novella
0.88
snippets
0.86
excerpts
0.84
recipes
0.83
highlights
0.83
gist
0.80
Cliffs
0.80
synopsis
0.80
POSITIVE LOGITS
survey
0.94
Survey
0.79
调查
0.77
Surveys
0.73
Comprehensive
0.73
Survey
0.73
survey
0.72
surveyed
0.72
調查
0.71
инде
0.69
Activations Density 0.033%