INDEX

Explanations

get up or move away

The neuron activates on small adverbial particles used in phrasal verbs (e.g. “up,” “off,” “down”).

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

not

-2.28

is

-1.91

 they

-1.86

 saying

-1.64

for

-1.64

 this

-1.63

 these

-1.59

say

-1.56

şi

-1.55

POSITIVE LOGITS

1.52

只会

1.46

腚

1.44

ajout

1.42

蕖

1.41

傒

1.39

 любых

1.38

嘻嘻

1.37

 frequentemente

1.37

prefer

1.35

Activations Density 0.006%