INDEX

Explanations

waiting until some condition

The neuron fires on the main content words—especially key action or status verbs—in a sentence (e.g. “answers,” “questions,” “delivered,” “fixed,” “obtained,” etc.).

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

as

-1.91

 only

-1.73

</i>

-1.28

 apenas

-1.23

 когда

-1.15

 μόνο

-1.15

 never

-1.14

}}/>

-1.13

 mostly

-1.13

当我

-1.11

POSITIVE LOGITS

 samarbe

1.34

Liar

1.18

渃

1.17

 baada

1.16

thée

1.16

ҕ

1.14

 samarbeid

1.14

 lámpara

1.13

 køb

1.12

 zijne

1.10

Activations Density 0.052%