INDEX

Explanations

murder warrant

The neuron activates specifically on the word “murder” (and its close forms) in the text.

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

。

-2.66



-2.59

didSet

-2.33



-2.19

生の

-2.17

)});

-2.16

)++;

-2.11

 bonté

-2.11

敉

-2.05

吖

-2.00

POSITIVE LOGITS

↵

4.50

3.38

 vektör

2.03

ೣ

2.00

 Grâce

2.00

 Beschluss

1.95

 nouveau

1.94

istiques

1.91

 cherished

1.90

 scenic

1.85

Activations Density 0.009%