INDEX

Explanations

drive-thru, drive safe, autonomous drive

The neuron detects the word “drive” (in any casing or as part of compound tokens).

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

tering

-1.03

ters

-0.90

 Rückseite

-0.87

tifies

-0.86

selling

-0.84

-0.82

some

-0.82

 collaborated

-0.81

 mães

-0.81

 Having

-0.79

POSITIVE LOGITS

 Drive

1.34

 drive

1.33

 DRIVE

1.23

Drive

1.10

DRIVE

1.07

drive

1.03

ways

1.03

 Drives

1.02

 drives

1.01

ться

0.94

Activations Density 0.014%