INDEX

Explanations

saw many changes

The neuron strongly activates on past-tense or past-participle forms of “see,” especially “saw” and “seen.”

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

 that

-2.33

on

-2.16

in

-1.88

 during

-1.77

 because

-1.64

 their

-1.63

if

-1.62

to

-1.59

 they

-1.57

 around

-1.54

POSITIVE LOGITS

 mélang

1.71

 celles

1.61

にします

1.60

offers

1.55

厹

1.55

 MULTIPLE

1.52

 élar

1.50

Antes

1.48

gives

1.47

反对

1.46

Activations Density 0.093%