INDEX

Explanations

This is/will

The neuron activates strongly on occurrences of the exact phrase “This Is” (as in section or article titles beginning with “This Is…”).

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

 terza

-0.87

volvement

-0.84

 شیائومی

-0.81

 picha

-0.81

طفال

-0.80

Berna

-0.77

第二

-0.77

 second

-0.76

érrez

-0.76

 fourth

-0.75

POSITIVE LOGITS

Is

1.06

is

1.06

 week

0.98

Next

0.91

Is

0.91

 That

0.90

黎明

0.89

That

0.89

 These

0.88

 Week

0.88

Activations Density 0.018%