INDEX

Explanations

until condition is met

This neuron detects verbs that describe attaining or hitting quantitative thresholds (e.g. "find," "reach," "become").

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

but

-1.55

 should

-1.52

as

-1.50

いますが

-1.48

 azonban

-1.45

 должны

-1.40

 then

-1.38

 хочет

-1.36

ınızı

-1.34

彼らは

-1.33

POSITIVE LOGITS

доста

1.44

thée

1.28

or

1.24

stache

1.24

்ட

1.23

༔

1.22

 вместе

1.21

 dolorosa

1.19

it

1.18

 where

1.18

Activations Density 0.050%