INDEX

Explanations

time to complete an action

The neuron fires on numeric expressions that denote lengths of time (e.g. “7 minutes,” “4 years,” “2–3 seconds,” “8 seconds,” 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

by

-1.45

 không

-1.28

氏は

-1.27

-1.25

了吗

-1.25

 only

-1.22

 только

-1.20

<i>

-1.20

↵↵

-1.19

 klasy

-1.15

POSITIVE LOGITS

 laim

1.54

 this

1.40

 viena

1.32

 bēr

1.28

籀

1.27

 syke

1.24

 moks

1.24

 lauk

1.23

 Lietu

1.23

 vienas

1.20

Activations Density 0.042%