INDEX

Explanations

numbers and specific labels

The neuron fires on verbs in the simple past tense (especially those ending in “-ed”).

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

you

-1.95

 your

-1.47

 चीज़ों

-1.30

 شما

-1.27

 nummer

-1.26

ṝ

-1.23

Ή

-1.23

wijl

-1.21

あなた

-1.21

 уже

-1.21

POSITIVE LOGITS

 again

1.70

 during

1.38

 took

1.34

 went

1.31

 today

1.30

to

1.30

 with

1.29

 різних

1.27

 allgemeinen

1.27

 gewöhnlich

1.25

Activations Density 0.005%