INDEX

Explanations

contrasting options

This neuron fires on instances of the negation “not.”

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

迎来

-1.28

Especificaciones

-1.27

ség

-1.20

🩻

-1.16

もありました

-1.16

 февра

-1.16



-1.13

 Vettel

-1.12

极其

-1.11

 mivel

-1.10

POSITIVE LOGITS

 just

1.82

 only

1.58

to

1.38

 merely

1.36

 because

1.34

 from

1.32

by

1.23

 только

1.18

 with

1.17

 without

1.16

Activations Density 0.014%