INDEX

Explanations

airport, boarding, or hotel locations

This neuron fires on location and directional terms in transport instructions—words like “near,” “to,” “located,” “in,” “departure,” “passengers,” etc., used when describing where to go or park.

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

-1.43

 this

-1.32

 galinha

-1.25

ษัท

-1.19

 сахара

-1.17

 forêts

-1.16

 wichtigsten

-1.14

𝑤

-1.13

做出

-1.12

-1.09

POSITIVE LOGITS

 dormitorios

1.71

 processos

1.41

 procedimentos

1.38

 esquí

1.38

 餐厅

1.35

 音乐

1.31

 trámites

1.29

 toallas

1.29

 baloncesto

1.25

 before

1.24

Activations Density 0.020%