INDEX

Explanations

knowing and not knowing

The neuron fires on occurrences of the verb “know,” especially in constructions like “didn’t know” or “little did I know.”

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

唔係

-1.16

 biztos

-1.13

 möglichst

-1.13

𝘞

-1.10

not

-1.09

 megjelen

-1.06

 solchen

-1.03

⬜

-1.02

 家用

-1.00

 schwarzen

-0.99

POSITIVE LOGITS

 until

2.45

how

2.16

或

1.66

or

1.63

或者

1.49

 about

1.48

 jusqu

1.48

 или

1.41

直到

1.37

 what

1.34

Activations Density 0.030%