INDEX

Explanations

towards

The neuron strongly activates on the directional preposition “towards.”

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

骘

-2.42

+"/"+

-2.42

-2.41

+="

-2.41

柃

-2.39

砗

-2.36

鶘

-2.36

ⴱ

-2.34

夤

-2.33

塒

-2.31

POSITIVE LOGITS

They

3.22

 Their

3.19

This

3.19

 They

3.14

 Truly

2.84

actéristi

2.83

 Firstly

2.75

かなり

2.72

There

2.70

 Totally

2.67

Activations Density 0.008%