INDEX

Explanations

veteran, building, Reporter, cout, Kill, concat

The neuron lights up on the word “veteran,” detecting mentions of “veteran” in text.

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

↵↵

-1.53

ных

-1.30

on

-1.27

ZAS

-1.24

↵↵↵

-1.24

 dessutom

-1.24

热水

-1.23

.}(

-1.22

 только

-1.16

 может

-1.16

POSITIVE LOGITS

if

1.52

返す

1.48

_,

1.36

the

1.36

*,

1.34

those

1.32

-,

1.31

digarh

1.30

initi

1.30

 apparente

1.30

Activations Density 0.147%