INDEX

Explanations

This is

The neuron primarily detects occurrences of the word “this” (the proximal demonstrative pronoun).

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

Akt

1.16

ης

1.12

นี่

1.09

นี

1.09

BAC

1.03

对外

1.03

 dimers

0.99

 нажмите

0.98

EZ

0.97

 Roop

0.97

POSITIVE LOGITS

 وهذه

1.89

 وهذا

1.51

guy

1.31

 particolare

1.27

at

1.27

pt

1.20

czas

1.13

ત્મ

1.12

 arada

1.10

ian

1.02

Activations Density 0.000%