INDEX

Explanations

paste it here

This neuron activates on deictic location words (e.g. “here,” “there,” “below,” “on”) that point to or situate content.

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

 here

1.55

 тут

1.48

 здесь

1.47

 هنا

1.43

 aquí

1.35

在這裡

1.34

 там

1.27

 aqui

1.27

 यहां

1.26

这里

1.23

POSITIVE LOGITS

abouts

1.52

ifrån

1.05

inafter

0.84

лі

0.73

ђе

0.67

fordshire

0.66

FORE

0.66

дика

0.64

ším

0.63

für

0.62

Activations Density 0.162%