INDEX

Explanations

as introducing examples

This neuron detects the phrase “such as,” i.e. the pattern “such” followed by “as” introducing examples.

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

Contacto

-1.59

所有的

-1.57

別人

-1.56

注意点

-1.55

每個

-1.51

Úl

-1.49

Ilustra

-1.48

..”

-1.48

整個

-1.45

無論

-1.45

POSITIVE LOGITS

for

2.50

 when

1.95

the

1.95

до

1.64

 that

1.63

from

1.59

 which

1.55

You

1.52

騭

1.48

 для

1.45

Activations Density 0.051%