INDEX

Explanations

descriptions of things

The neuron fires on temporal framing words and phrases that introduce or measure elapsed time (e.g. “In the past,” “For decades,” “Prior to,” etc.).

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

 currently

-1.75

 subsequently

-1.41

 presently

-1.38

 actualmente

-1.30

currently

-1.30

 છે

-1.27

sequently

-1.24

目前

-1.23

now

-1.23

Hiện

-1.23

POSITIVE LOGITS

was

1.74

といえば

1.73

と言えば

1.53

 only

1.40

 relied

1.33

 hanya

1.31

 وكان

1.31

 केवल

1.26

 chỉ

1.24

 ήταν

1.24

Activations Density 0.177%