INDEX

Explanations

polite interjections and statements

The neuron reacts to common conversational discourse markers and interjections (e.g. “Excuse me,” “trust me,” “mind you,” “guess what,” “believe it or not”) that signal asides or emphasis.

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

 Also

-1.91

 επίσης

-1.83

also

-1.66

 TAMBÉM

-1.63

 também

-1.61

 också

-1.58

 Moreover

-1.56

 ALSO

-1.56

 citazioni

-1.54

 This

-1.53

POSITIVE LOGITS

the

2.00

’

1.90

 городов

1.62

 what

1.52

最后的

1.50

it

1.50

him

1.48

秾

1.47

 reputed

1.44

 görüş

1.41

Activations Density 0.023%