INDEX

Explanations

future intention or state

The neuron fires strongly on verbs and auxiliaries that mark intentions, requests, or future actions (e.g. going, will, asking, looking for).

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

resave

-1.42

既に

-1.34

ÓD

-1.32

 тільки

-1.32

 その

-1.30

 sadece

-1.28

橢

-1.27

 только

-1.27

一方で

-1.24

喜んで

-1.20

POSITIVE LOGITS

now

1.63

now

1.61

 devenu

1.45

蚰

1.36

 moeten

1.25

蠃

1.24

 myself

1.21

༵

1.18

 verschillen

1.16

gged

1.15

Activations Density 0.189%