INDEX

Explanations

getting code, file, or data

The neuron detects the occurrence of the verb “get” (in any capitalization) in code or comments.

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

 diese

-1.39

↵

-1.27

 enfermera

-1.26

 одного

-1.23

缯

-1.21

 diesen

-1.20

これは

-1.16

ógio

-1.15

we

-1.11

 permanente

-1.10

POSITIVE LOGITS

 from

2.86

and

1.67

the

1.41

 their

1.36

 information

1.34

nější

1.30

 procurer

1.27

rid

1.24

 what

1.23

さんから

1.20

Activations Density 0.068%