INDEX

Explanations

casting to

This neuron detects the substring “cast” (e.g. in “cast,” “Cast,” “CastException,” etc.), i.e. references to casting operations.

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

脔

-2.08

 regulación

-1.88

苼

-1.86

艄

-1.82

unpublished

-1.68

 motores

-1.66

跸

-1.66

 abbiamo

-1.65

 них

-1.64

 restauración

-1.64

POSITIVE LOGITS

to

2.28

”,

1.75

 意思

1.74

—

1.74

1.73

↵

1.71

</

1.66

For

1.59

’

1.56

 There

1.55

Activations Density 0.010%