INDEX

Explanations

existence and availability information

The neuron responds to words that describe the presence, absence, or availability status of something (e.g. “lack,” “still active,” “available,” “dead,” “availability”).

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

 three

-1.35

我们

-1.28

or

-1.25

５

-1.21

或

-1.21

Ecotoxicity

-1.19

 five

-1.17

使用

-1.17

 last

-1.16

aim

-1.16

POSITIVE LOGITS

 such

1.27

olí

1.24

utí

1.23

 reeks

1.21

imantan

1.21

uatu

1.20

ressum

1.20

ensione

1.18

 люб

1.18

ليات

1.17

Activations Density 0.150%