INDEX

Explanations

create the

The neuron fires on instructions about creating or constructing something, especially the verb “create” (and its variants) introducing an item or step.

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

vertible

-1.84

quisition

-1.80

禵

-1.77

🫃

-1.76

 ludzi

-1.73

 ЛИ

-1.73

 Traducción

-1.72

ᐕ

-1.69

especially

-1.66

 Seleccione

-1.66

POSITIVE LOGITS

2.92

an

2.14

‘

1.91

If

1.71

.”

1.70

“

1.59

 which

1.57

The

1.55

 necessary

1.53

 необходи

1.51

Activations Density 0.105%