INDEX

Explanations

leaving behind or not taking

This neuron detects occurrences of the verb “leave” (and its inflected forms such as “leaves” or “left”).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

िया

1.63

iezan

1.45

おります

1.36

 apertura

1.35

ian

1.27

কেন্দ্র

1.27

लेकिन

1.26

ünün

1.23

하지만

1.22

 américains

1.21

POSITIVE LOGITS

م

1.54

taker

1.44

 cobert

1.36

sine

1.36

srv

1.30

لها

1.30

σιμοποι

1.30

 преде

1.30

вт

1.27

 plow

1.27

Activations Density 0.036%