INDEX

Explanations

specific names and units

The neuron strongly activates on tokens representing units of measure (e.g. “meters,” “liters,” “microseconds,” “kilograms,” etc.) in conversion‐style questions.

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

וּ

-1.32

isait

-1.23

 mög

-1.23

 kön

-1.19

 activities

-1.18

 cosméticos

-1.16

 Controle

-1.12

ynchronously

-1.10

álise

-1.10

 activity

-1.10

POSITIVE LOGITS

 esimerkiksi

1.27

iddhar

1.26

ὺς

1.20

めます

1.19

ଷ

1.17

 laajenta

1.15

 ویتامین

1.13

 nyky

1.12

Gambas

1.10

തു

1.07

Activations Density 0.008%