INDEX

Explanations

shown in Figure

The neuron responds to references to figures or illustrations (e.g. words like “sketched,” “shown,” “depicted,” and “in Fig.” alongside figure numbers).

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

 chỉ

-1.14

就跟

-1.11

 apenas

-1.07

≠

-1.04

only

-1.04

 .......

-1.01

 corresponde

-0.95

只剩下

-0.94

 anderen

-0.94

就把

-0.93

POSITIVE LOGITS

in

1.55

 below

1.46

 Figure

1.23

Figure

1.16

maktadır

1.14

on

1.12

したのが

1.10

 Ocak

1.09

图

1.06

 Table

1.05

Activations Density 0.026%