INDEX

Explanations

phonemes, words, and lexical units

This neuron flags occurrences of the terms “phone,” “phoneme,” or “phones” (i.e. references to phonetic units).

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

otriva

-0.75

textwidth

-0.71

 preſ

-0.71

acting

-0.69

Fa

-0.69

 webbplats

-0.69

pont

-0.68

రం

-0.67

atorics

-0.67

الر

-0.66

POSITIVE LOGITS

 morphe

1.59

 words

1.40

 units

1.23

lex

1.15

 sign

1.12

 elements

1.02

 signs

1.02

 segments

1.01

 lexical

1.00

 лек

1.00

Activations Density 0.042%