INDEX

Explanations

determiners and pronouns

The neuron strongly activates on personal pronouns (object and possessive forms) such as “our,” “us,” “its,” and “them.”

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

⁃

-1.23

⅗

-1.19

siamo

-1.19

 دیگری

-1.16

aise

-1.14

 daquele

-1.13

сию

-1.11

 berech

-1.10

และ

-1.09

 later

-1.08

POSITIVE LOGITS

 only

1.87

 this

1.67

 with

1.26

 already

1.25

 ONLY

1.23

 sólo

1.23

fourths

1.17

 única

1.15

 только

1.15

 about

1.15

Activations Density 0.222%