INDEX

Explanations

location or direction adverbs

The neuron fires most strongly on the key content word (usually a noun or verb) at the end of a clause or sentence.

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

-3.22

 והוא

-1.18

 tazas

-1.12

atürk

-1.10

 ejecut

-1.05

ärsk

-1.04

ukunft

-1.02

ciąż

-1.01

xxi

-0.99

ようで

-0.99

POSITIVE LOGITS

 accordingly

1.48

 therefrom

1.30

 then

1.28

 while

1.26

 there

1.26

ativis

1.15

for

1.13

 instead

1.12

 inserire

1.10

 what

1.09

Activations Density 0.090%