INDEX

Explanations

in the sense that/of

The neuron activates on occurrences of the word “sense,” especially in formulaic phrases like “in the sense of” or “in the sense that.”

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

訁

-1.85

 inserir

-1.66

掼

-1.57

美国

-1.54

美國

-1.48

神器

-1.48

 reservados

-1.47

关于

-1.47

 ويكي

-1.45

 белом

-1.43

POSITIVE LOGITS

 that

3.44

</h3>

2.23

er

1.91

1.84

of

1.65

1.64

able

1.62

—

1.53

 jeweils

1.48

しっかりと

1.47

Activations Density 0.006%