INDEX

Explanations

all-around

The neuron activates on occurrences of the word “all,” especially when used in compound phrases (e.g. “All-Wheel,” “all-rounder,” “All-Purpose,” “All-in-One”).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

לון

-0.93

全て

-0.89

 всю

-0.87

 Unterscheidung

-0.85

 একটি

-0.82

ทั้งหมด

-0.82

룽

-0.81

 всички

-0.80

整个

-0.79

CARA

-0.79

POSITIVE LOGITS

you

1.85

 ages

1.58

 seasons

1.55

 kinds

1.46

 around

1.40

 types

1.39

 aspects

1.30

round

1.25

 year

1.25

age

1.24

Activations Density 0.033%