INDEX

Explanations

connections between things

The neuron triggers on linking words that signal relationships between things—especially prepositions (e.g. “between,” “to,” “on,” “for”) and related link-forming terms.

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

 which

-2.36

 while

-1.97

all

-1.92

 just

-1.92

 provide

-1.91

 providing

-1.89

 that

-1.89

 time

-1.80

 still

-1.80

 this

-1.80

POSITIVE LOGITS

 También

1.96

扨

1.94

 horrid

1.89

 iſt

1.88

 goofy

1.85

 delightfully

1.85

 deliciously

1.79

 Проци

1.78

 recensioni

1.78

 två

1.77

Activations Density 0.143%