INDEX

Explanations

in the presence of

The neuron detects occurrences of the word “presence,” especially in the phrase “in the presence of.”

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

-2.63

᯾

-2.11

詎

-2.05

ományos

-2.03

 verfügb

-1.99

ेत्र

-1.97

 podemos

-1.93

蟥

-1.92

},

-1.88

 ショート

-1.88

POSITIVE LOGITS

Недостатки

2.11

Выводы

2.09

鶿

2.06

Ито

2.02

ญี่ป

2.02

مقدمه

1.99

但是

1.98

↵

1.98

 köy

1.98

Особенности

1.95

Activations Density 0.014%