INDEX

Explanations

particular instances; specific types

The neuron strongly activates on the word “particular” (marking instances of that specific-descriptor adjective).

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

*"

-1.65

*'

-1.63

-1.50

he

-1.44

 он

-1.37

*',

-1.37

roughly

-1.36

↵↵↵

-1.32

The

-1.31

*",

-1.30

POSITIVE LOGITS

 these

2.23

 certain

2.19

 such

1.99

 those

1.83

ógico

1.81

 sections

1.80

喽

1.77

lawesi

1.76

ógicos

1.72

 parts

1.70

Activations Density 0.056%