INDEX

Explanations

more or less quantities

The neuron activates on words that express quantitative change or comparison—especially comparative or relative-amount terms (e.g. higher, lower, fewer, more, decline, mean).

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

it

-1.67

nachron

-1.33

the

-1.28

 there

-1.20

AfterViewInit

-1.17

its

-1.16

‘

-1.15

-1.13

 normaal

-1.11

Hæ

-1.11

POSITIVE LOGITS

for

2.34

ερ

1.40

orescent

1.34

Karakteristik

1.33

 mať

1.33

 জন্য

1.33

ícia

1.31

เรา

1.30

któ

1.28

 něj

1.27

Activations Density 0.037%