INDEX

Explanations

lack of or absence

The neuron activates on words marking causal or passive-voice relationships (e.g. “because,” “due,” “to,” “of,” “by”)—i.e. connective prepositions in phrases like “due to” or “because of.”

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

 hecta

-1.22

pè

-1.21

ene

-1.21

 effici

-1.18

 substan

-1.13

 deri

-1.13

 tranquill

-1.12

 momen

-1.12

dè

-1.12

 dégustation

-1.10

POSITIVE LOGITS

or

3.27

 either

2.95

 либо

2.20

either

2.20

Either

1.86

 entweder

1.80

 perhaps

1.79

 Either

1.77

 various

1.76

 или

1.70

Activations Density 0.070%