INDEX

Explanations

numbers followed by superscripts or indices

The neuron responds whenever it sees standalone digits or numeric indices in mathematical notation (e.g. subscripts or exponents).

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

to

-2.16

 они

-2.06

it

-2.00

by

-1.97

 this

-1.91

 from

-1.90

,”

-1.87

as

-1.79

 mereka

-1.73

↵↵

-1.72

POSITIVE LOGITS

2.50

 hordes

2.00

 craz

1.90

 hinders

1.89

 jarring

1.87

1.83

胗

1.80

神

1.77

又一次

1.76

OUTUBE

1.73

Activations Density 0.012%