INDEX

Explanations

the introducing phrase

The neuron strongly activates on sentence-initial words (especially common pronouns and articles like “It,” “The,” “If,” “Here,” etc.).

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

it

-2.06

we

-1.86

you

-1.82

なんですよ

-1.77

 this

-1.73

 bebés

-1.63

 they

-1.63

 there

-1.61

も行

-1.57

preso

-1.52

POSITIVE LOGITS

</b>

1.82

“

1.77

breviations

1.76

 البته

1.68

'';

1.61

了一会儿

1.60

ba

1.55

1.53

ος

1.53

 ‘‘

1.52

Activations Density 0.063%