INDEX

Explanations

The neuron detects occurrences of the word “else.”

someone else

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

and

-3.33

archiwizowane

-2.42

鰾

-2.34

 efekt

-2.33

 véritable

-2.16

蜴

-2.05

 手绘

-1.98

퐹

-1.97

Ꮞ

-1.96

 短发

-1.95

POSITIVE LOGITS

は

2.33

та

2.09

2.02

matchCondition

1.97

there

1.96

1.94

 italianos

1.94

=\""

1.93

дев

1.90

."

1.89

Activations Density 0.005%