INDEX

Explanations

following sentence start

The neuron is primarily firing on sentence-terminating punctuation (full stops) and other strong boundary markers, i.e. it detects the ends of sentences.

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

 also

-1.31

 other

-0.94

 भी

-0.90

なども

-0.90

也有

-0.88

all

-0.84

盥

-0.83

 Также

-0.80

 spørs

-0.79

 заканчи

-0.78

POSITIVE LOGITS

 avulla

1.29

σιο

1.04

 sayesinde

1.02

ଵ

0.98

iaire

0.98

jį

0.96

以内

0.95

 chúng

0.95

 εφαρ

0.95

onesa

0.94

Activations Density 0.096%