INDEX

Explanations

it all followed by a concluding verb

This neuron detects discourse-level intensifier/adverb tokens (e.g. “all,” “just,” “works fine,” etc.) that serve as filler or emphasis in a sentence.

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

猛然

-1.45

logitech

-1.44

fatter

-1.41

-1.37

猛地

-1.37

is

-1.35

 этот

-1.34

atguigu

-1.34

mycompany

-1.32

 alcun

-1.30

POSITIVE LOGITS

 zuge

1.59

 culmin

1.53

 happening

1.47

dem

1.38

一个小

1.37

 cómo

1.33

aé

1.30

++]=

1.30

{$\

1.27

"},

1.26

Activations Density 0.017%