INDEX

Explanations

facebook mentions

The neuron fires on any occurrence or reference to “Facebook” (including in URLs).

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

了

-1.83

]^

-1.67

меся

-1.59

cture

-1.55

)-\

-1.54

口气

-1.54

匍

-1.53

колко

-1.52

</sup>

-1.51

 снимка

-1.48

POSITIVE LOGITS

茈

1.73

ஂ

1.63

大学の

1.60

 первом

1.52

riculares

1.51

him

1.47

幛

1.43

Facebook

1.41

 negras

1.37

镳

1.36

Activations Density 0.040%