INDEX

Explanations

The neuron fires strongly on occurrences of the word “every” (and its capitalized form) in text.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ী

1.45

 hoher

1.41

४

1.31

luz

1.30

 critérios

1.27

шены

1.27

 воздей

1.26

новых

1.26

 folhas

1.25

ных

1.24

POSITIVE LOGITS

 conceivable

1.44

 every

1.39

 imaginable

1.35

 türlü

1.24

every

1.21

 ounce

1.20

 EVERY

1.20

 وكل

1.20

 member

1.17

 Every

1.17

Activations Density 0.053%