INDEX

Explanations

usher leading or initiating

The neuron detects occurrences of the substring “usher” in any context.

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

橚

-2.92

袛

-2.70

ſſen

-2.63

投资者

-2.59

鸲

-2.58

 می‌کند

-2.47

⎘

-2.44

姈

-2.42

 orígenes

-2.38

補足

-2.36

POSITIVE LOGITS

 Then

3.30

 What

2.78

2.61

2.53

</td>

2.52

 This

2.39

</i>

2.36

 usher

2.34

<i>

2.33

 There

2.23

Activations Density 0.002%