INDEX

Explanations

exit, status, age, route

This neuron detects numeric tokens (numbers and numeric measurements).

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

 this

-0.96

 before

-0.84

 time

-0.82

场比赛

-0.81

、“

-0.81

我们可以

-0.81

 dourada

-0.79

 menurun

-0.79

Before

-0.79

 because

-0.79

POSITIVE LOGITS

 ellos

0.87

 AMAZING

0.85

eacute

0.85

 geprüft

0.85

 Duitse

0.84

 distinguer

0.84

 ahora

0.83

 papild

0.82

海道

0.81

 akci

0.80

Activations Density 0.207%