INDEX

Explanations

too [adjective/adverb] to [verb]

This neuron detects constructions where an adjective or adverb is followed by the infinitive marker “to” (e.g. “too … to,” “early … to,” “weak … to”).

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

Sabes

-1.20

哐

-1.17

genomen

-1.17

 experto

-1.11

Ál

-1.11

一樣

-1.09

zake

-1.09

pravy

-1.08

Дизайн

-1.06

季節

-1.05

POSITIVE LOGITS

to

1.91

be

1.42

1.39

1.35

 that

1.34

 than

1.33

go

1.29

到底是

1.27

是怎么回事

1.24

 antes

1.23

Activations Density 0.025%