INDEX

Explanations

accepted and obvious

The neuron fires on occurrences of the phrase “take for granted.”

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

-1.75

-1.73

</b>

-1.70

�

-1.66

-1.55

 endeavors

-1.45

parseFrom

-1.43

⩽

-1.43

 craindre

-1.42

the

-1.38

POSITIVE LOGITS

 that

2.41

For

1.59

This

1.57

錞

1.52

 gial

1.52

élève

1.51

 NSMutable

1.50

breviations

1.50

itattu

1.47

ológicos

1.45

Activations Density 0.020%