INDEX

Explanations

more information

The neuron activates on hyperlinks or pointers to external resources (e.g. URLs and “see/visit” link cues).

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

Crítica

-1.33

茳

-1.31

 desastre

-1.27

 également

-1.25

UTELY

-1.22

日閲覧

-1.20

 Solutions

-1.20

ığınız

-1.20

ADUATE

-1.18

>::

-1.17

POSITIVE LOGITS

 than

2.17

 poszczegól

1.40

of

1.32

 Estructura

1.32

quise

1.28

 SEÑ

1.27

 from

1.27

how

1.26

 navegador

1.23

Activations Density 0.041%