INDEX

Explanations

actions causing movement or change

The neuron fires on vivid sensory or physical action words—especially dynamic verbs describing things moving, flowing, dripping, crunching, etc.

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

着一个

-1.40

 this

-1.28

who

-1.27

 Segel

-1.26

lahraga

-1.26

 kablo

-1.24

 また

-1.18

now

-1.14

if

-1.14

it

-1.13

POSITIVE LOGITS

 desde

1.51

 pierwszym

1.40

 into

1.33

észetes

1.32

1.26

 colère

1.23

裡的

1.23

 แต่

1.22

 through

1.21

 zakoń

1.20

Activations Density 0.027%