INDEX

Explanations

star rating

This neuron responds to the numeric value in “star‐rating” expressions (i.e. the digits that follow “star” in a star rating).

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

ción

-1.86

盦

-1.73

 studenten

-1.69

緖

-1.66

 kolorze

-1.65

 גם

-1.60

着的

-1.60

тин

-1.60

tldr

-1.60

 också

-1.59

POSITIVE LOGITS

':

2.17

2.06

my

1.96

?”

1.94

}=

1.87

}$,

1.77

1.76

 vergangenen

1.73

are

1.71

 спросил

1.69

Activations Density 0.027%