INDEX

Explanations

cache memory

The neuron activates on mentions of computer memory types—especially cache and flash/SRAM memory.

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

 godk

-1.31

 Longueur

-1.28

真的

-1.27

 Améli

-1.27

 autorización

-1.27

-1.25

 öns

-1.24

 ਲਈ

-1.23

extField

-1.23

 necessários

-1.23

POSITIVE LOGITS

 να

1.41

out

1.25

 commonplace

1.12

 cache

1.09

ヤン

1.07

 Your

1.07

を発表

1.07

 στα

1.06

]$.

1.05

 bâtiments

1.05

Activations Density 0.072%