INDEX

Explanations

lead to / has to

The neuron fires strongly on common English verbs and auxiliaries (e.g. “get,” “have,” “will,” “takes,” “does,” “think,” “would”).

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

 there

-1.86

 another

-1.25

 that

-1.22

 once

-1.18

 paisagem

-1.17

 Ayuntamiento

-1.16

ragamo

-1.15

let

-1.14

 bouteille

-1.14

 tornou

-1.13

POSITIVE LOGITS

[:,:,

1.27

 arī

1.18

fVar

1.15

Elke

1.15

❓

1.14

ität

1.14

 ろう

1.13

🍈

1.13

ará

1.13

 даже

1.11

Activations Density 0.126%