INDEX

Explanations

phrases separated by commas

This neuron detects emphatic three‐time repetitions of the same word (e.g. “X, X, X”) 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

很久

-1.58

了一

-1.54

趼

-1.37

-1.32

很多人

-1.30

hichever

-1.27

 demonio

-1.26

Religion

-1.25

gesteld

-1.24

 engend

-1.22

POSITIVE LOGITS

されていた

1.70

</em>

1.49

<>());

1.48

)?;

1.45

','

1.43

],

1.43

yelitis

1.42

mybatisplus

1.41

 sowie

1.40

 trouvait

1.40

Activations Density 0.017%