INDEX

Explanations

comes in, comes up, comes with

The neuron strongly responds to occurrences of the verb “come” (in its various inflected forms, e.g. come, came, comes).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

ات

1.63

ي

1.49

И

1.45

1.43

О

1.32

את

1.31

ি

1.30

К

1.30

І

1.27

ээр

1.26

POSITIVE LOGITS

prising

1.17

ptions

1.14

ên

1.09

or

1.05

 Medicines

1.02

りの

1.01

 Voluntary

1.01

ตรฐาน

0.98

 나타

0.98

Activations Density 0.098%