INDEX

Explanations

placeholder

The neuron specifically activates on occurrences of the word “placeholder.”

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

 There

-2.70

 verdaderas

-2.42

 vostre

-2.31

RTCK

-2.30

 แต่

-2.17

אלי

-2.16

 articoli

-2.14

导演

-2.13

籥

-2.11

there

-2.06

POSITIVE LOGITS

姈

3.11

淠

2.86

↵

2.83

2.63

Ꭳ

2.52

<bos>

2.44

_{

2.31

趼

2.30

2.25

言うと

2.23

Activations Density 0.001%