INDEX

Explanations

part of a category

The neuron activates on numeric tokens, i.e. numbers and digits representing quantities.

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

 through

-1.20

its

-1.12

via

-1.11

ございました

-1.02

Attached

-1.00

 adecuada

-0.99

𝘪

-0.98

cipta

-0.93

 aquifers

-0.93

 while

-0.92

POSITIVE LOGITS

 which

1.82

 that

1.73

 που

1.45

 الذي

1.30

 других

1.24

 autres

1.23

 که

1.19

ซึ่ง

1.17

 other

1.16

 monit

1.11

Activations Density 0.117%