INDEX

Explanations

proactive projects

The neuron activates strongly on occurrences of the word “initiative” (and its plural) in the text.

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

-3.17

雱

-2.91

we

-2.83

鬭

-2.63

 Geschä

-2.25

 kleines

-2.23

 berühm

-2.23

.;

-2.17

 Produkten

-2.14

 gänzlich

-2.14

POSITIVE LOGITS

咈

2.97

!”

2.63

纈

2.59

錒

2.56

፨

2.53

辚

2.45

 ſou

2.44

⤞

2.42

setBold

2.42

呖

2.38

Activations Density 0.009%