INDEX

Explanations

arrival

The neuron strongly fires on past‐tense narrative indicators—especially verbs like "arrived," "pulled," "got," etc.—marking key story events.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

去看

0.78

回去

0.74

去

0.68

gå

0.62

外出

0.60

 Exports

0.59

要去

0.59

 Export

0.59

出去

0.58

 exports

0.58

POSITIVE LOGITS

 arrived

2.72

 arrive

2.69

 arriving

2.66

 arrival

2.58

 arrives

2.58

 arriv

2.48

arrive

2.41

 Arrival

2.39

 도착

2.36

Arrival

2.34

Activations Density 0.462%