INDEX

Explanations

never forget

The neuron strongly activates on words and phrases about recalling or failing to recall things (e.g. “remember,” “forget,” “never forget,” “remembers”).

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

ждения

-1.77

cadilly

-1.53

 サンダル

-1.51

-1.44

总是

-1.41

:/

-1.40

 ۱۲

-1.38

を使用

-1.36

 علاوه

-1.36

 largely

-1.35

POSITIVE LOGITS

 from

2.56

and

1.91

 because

1.91

how

1.80

 took

1.76

 when

1.73

 went

1.69

or

1.65

two

1.65

 during

1.58

Activations Density 0.012%