INDEX

Explanations

this followed by a noun

The neuron responds to verbs that describe doing or fixing something (e.g. “correct,” “reproducing,” “transform”)—i.e. procedural‐action verbs.

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

and

-1.76

'./

-1.73

you

-1.66

 honom

-1.62

 ruedas

-1.60

biendo

-1.60

りましたが

-1.59

 nawet

-1.57

aarrggbb

-1.55

ohjel

-1.55

POSITIVE LOGITS

in

2.22

on

2.16

for

2.08

 then

1.73

alians

1.70

 with

1.69

 على

1.63

 на

1.62

</em>

1.56

 same

1.53

Activations Density 0.119%