INDEX

Explanations

describes arrangement or characteristics

The neuron fires on dynamic “coverage” verbs and related words—terms like line, dot, fill, dominate, characterizes, define—that describe something being laid out across, covering, or defining an area.

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

 similar

-1.61

 realizado

-1.36

ですか

-1.34

zień

-1.30

altor

-1.29

larıyla

-1.27

-1.23

eddable

-1.21

 dadurch

-1.20

mentado

-1.16

POSITIVE LOGITS

our

2.39

the

2.02

 them

1.59

 their

1.55

مقال

1.51

Tetapi

1.45

 tchèque

1.45

 nhằm

1.45

 możemy

1.41

 нашего

1.38

Activations Density 0.083%