INDEX

Explanations

knows or realizes things

The neuron fires on verbs expressing knowledge or perception (e.g. know, knew, known, looks, realised).

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

 from

-1.76

 with

-1.72

 when

-1.63

 after

-1.49

 Here

-1.42

 section

-1.41

 version

-1.38

at

-1.34

one

-1.33

 From

-1.32

POSITIVE LOGITS

컹

1.40

lenmiş

1.38

抿

1.32

Книга

1.30

 hecta

1.30

 två

1.30

鱚

1.29

 île

1.29

どれも

1.27



1.26

Activations Density 0.047%