INDEX

Explanations

presence or attendance

Explanation of neuron 4 behavior: the main thing this neuron does is find mentions of being present or in attendance.

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

 baada

-0.91

 icona

-0.91

 után

-0.89

してから

-0.86

 attesa

-0.85

 μέσα

-0.84

 pozwala

-0.83

juje

-0.82

 înainte

-0.82

 tìm

-0.81

POSITIVE LOGITS

to

1.46

at

1.33

全程

1.08

giày

1.05

for

1.04

Какво

1.03

 during

0.96

 witness

0.92

َه

0.91

োগ

0.90

Activations Density 0.036%