INDEX

Explanations

by analogy with

The neuron responds to abstract relational or connective terms (e.g. “analogy,” “association,” “correlative,” “accordingly”) that link or compare concepts.

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

 цвят

-1.21

 actor

-1.19

 then

-1.18

 hergestellt

-1.16

チェンジ

-1.15

战队

-1.13

 крайней

-1.12

してみる

-1.11

</em>

-1.09

 colgantes

-1.09

POSITIVE LOGITS

我们

1.59

 ſou

1.31

芣

1.21

some

1.18

vanju

1.17

istung

1.16

enskap

1.16

posé

1.15

 certaines

1.14

жного

1.14

Activations Density 0.086%