INDEX

Explanations

code imports and definitions

The neuron fires most strongly on code tokens containing underscores (i.e. snake_case identifiers).

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

for

-1.41

 testGet

-1.30

at

-1.29

chränkungen

-1.23

 over

-1.22

on

-1.21

jsonwebtoken

-1.17

赑

-1.17

 entanto

-1.16

 yılında

-1.16

POSITIVE LOGITS

my

1.47

 lately

1.37

čeno

1.34

 ワゴン

1.26

了我

1.26

due

1.25

thanks

1.24

 perchè

1.23

yeah

1.20

 grazie

1.20

Activations Density 0.051%