INDEX

Explanations

intended to be

The neuron detects words and phrases that indicate scheduled or planned future events (e.g. “scheduled,” “to be,” “supposed to,” “planned to,” and associated dates or times).

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

 próprias

-1.63

łbym

-1.49

 gelegt

-1.42

 schlä

-1.42

crí

-1.41

dición

-1.41

 mám

-1.40

 первом

-1.39

 것입니다

-1.37

 maneiras

-1.35

POSITIVE LOGITS

of

1.66

if

1.45

be

1.42

 able

1.36

 tetapi

1.32

 Universität

1.26

 start

1.26

at

1.25

ajte

1.21

Activations Density 0.034%