INDEX

Explanations

judgment and personal reaction

The neuron flags self‐referential and directly addressed language—first‐ and second‐person pronouns and phrases (I, we, you, our, my) that convey personal/opinionated stance.

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

 including

-1.47

↵↵↵↵

-1.44

 include

-1.41

↵↵↵

-1.38

 work

-1.35

 employee

-1.28

 wraz

-1.23

番外

-1.23

 wise

-1.21

↵

-1.20

POSITIVE LOGITS

 liknande

1.56

你现在

1.45

 Вода

1.41

Теперь

1.30

因为它

1.24

 semillas

1.24

涐

1.23

Бележки

1.21

نب

1.21

อ่าน

1.21

Activations Density 0.170%