INDEX

Explanations

face and its appearance or covering

The neuron strongly activates on occurrences of the word “face” (and its plural “faces”), i.e. when the text refers to faces or facial covering.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

).

-1.82

),

-1.78

);

-1.73

 builds

-1.68

-1.65

的确

-1.63

ですので

-1.60

 browns

-1.56

na

-1.55

 utilizarse

-1.54

POSITIVE LOGITS

または

2.19

 یا

2.08

และ

1.97

 или

1.95

㍑

1.88

𖣘

1.84

或

1.82

絝

1.77

特斯拉

1.73

colgante

1.71

Activations Density 0.029%