INDEX

Explanations

words often followed by punctuation

the neuron is keyed to content‐bearing nouns—especially topic or object nouns serving as titles or key subjects in the text.

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

 from

-1.69

if

-1.63

mschrijving

-1.52

typal

-1.52

parral

-1.51

 менее

-1.50

MAXN

-1.45

 caoutchouc

-1.41

*/

-1.40

 which

-1.40

POSITIVE LOGITS

它们

1.62

 réjou

1.58

 seseorang

1.57

 estrena

1.55

 soutě

1.55

 verschillende

1.54

慫

1.52

 anuncia

1.51

粜

1.50

凄く

1.48

Activations Density 0.204%