INDEX

Explanations

bunny ears or shaped bunny

The neuron is keyed to occurrences of the word “bunny” (in various contexts and forms).

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

“

-2.84

。“

-2.84

(“

-2.67

㌘

-2.63

.”

-2.61

衚

-2.58

，“

-2.55

↵

-2.53

 gesamte

-2.52

、“

-2.48

POSITIVE LOGITS

<i>

3.48

 When

2.66

 Когда

2.61

</b>

2.53

Those

2.45

 resolute

2.38

ヴァン

2.25

<b>

2.23

</h1>

2.20

Activations Density 0.002%