INDEX

Explanations

integration with

The neuron fires on words and phrases describing “integration” (e.g. integrate, integrating, integration) between software components or services.

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

第七

-1.52

第六

-1.52

：「

-1.41

この

-1.37

addCriterion

-1.33

 hän

-1.32

نظ

-1.31

 srebr

-1.31

 laatste

-1.30

weilen

-1.30

POSITIVE LOGITS

 with

1.45

 from

1.40

 手機

1.38

of

1.36

1.35

from

1.29

 respective

1.29

1.27

nitř

1.25

 传统

1.24

Activations Density 0.047%