INDEX

Explanations

encouraging activities and outcomes

The neuron strongly activates on occurrences of the verb “encourage.”

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

-3.30

𖣘

-2.58

栳

-2.47

-2.41

 reun

-2.27

we

-2.22

樒

-2.20

 Ereignisse

-2.14

 vivía

-2.13

觧

-2.11

POSITIVE LOGITS

哢

2.50

’,

2.42

[]:

2.22

’.

2.19

ведення

2.13

2.11

providedIn

2.09

ússia

2.08

):

2.06

водит

2.06

Activations Density 0.003%