INDEX

Explanations

some examples are:

The neuron detects words and expressions that introduce lists or categories (e.g. list-heading and enumeration cues).

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

GRAFIA

-1.65

 puisque

-1.34

GRAFÍA

-1.33

 начинают

-1.32

ammlung

-1.28

 flot

-1.27

 analyser

-1.27

 บาท

-1.25

 '/';

-1.24

着一

-1.23

POSITIVE LOGITS

 that

1.58

of

1.55

 människor

1.43

how

1.42

 these

1.41

 durchführen

1.40

 from

1.37

 which

1.35

 also

1.34

 bahkan

1.34

Activations Density 0.047%