INDEX

Explanations

all in all

This neuron detects summary or concluding transition phrases (e.g. “all in all,” “overall,” “on the whole”).

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

or

-1.66

 several

-1.45

 with

-1.44

and

-1.41

at

-1.39

in

-1.35

 without

-1.33

one

-1.32

 after

-1.23

on

-1.22

POSITIVE LOGITS

1.41

 insgesamt

1.30

 aparência

1.28

 lavable

1.27

 arquitetura

1.20

云

1.14

 rotina

1.13

 thì

1.13

 perpé

1.11

 comunque

1.10

Activations Density 0.012%