INDEX

Explanations

goal or objective

The neuron detects occurrences of explicit goal/objective statements (the word “goal” introducing project or study aims).

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

 those

-2.17

the

-2.11

 other

-1.77

 preparando

-1.68

ními

-1.63

...'

-1.63

 butterknife

-1.61

nými

-1.61

 cejas

-1.59

 comprando

-1.59

POSITIVE LOGITS

of

2.59

is

1.95

 From

1.71

 после

1.65

1.54

 According

1.52

 jedną

1.52

 There

1.52

</b>

1.50

но

1.48

Activations Density 0.022%