INDEX

Explanations

US state government administration

The neuron specifically fires on occurrences of the word “State,” especially when it appears as part of a proper name (e.g. “Washington State,” “New York State”).

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

-2.11

-2.05

-1.87

-1.80

-1.77

</strong>

-1.76

ys

-1.70

thée

-1.67

洐

-1.66

นี้

-1.66

POSITIVE LOGITS

 verheir

2.09

 ofreci

2.02

 которы

1.91

 verheiratet

1.84

ارتمان

1.80

are

1.76

 když

1.76

仉

1.75

不

1.73

 nicht

1.70

Activations Density 0.007%