INDEX

Explanations

traveling to places

The neuron fires on mentions of traveling, i.e. the verb “travel” in its various inflected forms.

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

 walk

-0.99

 walking

-0.98

 iliy

-0.96



-0.93

 carcasa

-0.93



-0.92

 Leyden

-0.90

 flurry

-0.90



-0.90

 strolling

-0.90

POSITIVE LOGITS

to

2.88

 through

1.45

 from

1.44

 around

1.42

 with

1.40

 dieną

1.18

𓋼

1.13

 abroad

1.10

 distances

1.08

 across

1.07

Activations Density 0.016%