INDEX

Explanations

ODE of

The neuron is activated by numeric tokens containing decimal points (i.e. floating-point numbers).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

也将

0.74

 resided

0.71

 మరింత

0.71

 उन्होंने

0.70

 corresponded

0.69

 сега

0.68

 주요

0.66

 appeared

0.66

主要的

0.65

还将

0.63

POSITIVE LOGITS

 бывает

1.53

 бывают

1.43

 সাধারণত

1.32

 suele

1.27

こういう

1.25

when

1.17

 suelen

1.16

一般来说

1.14

 όταν

1.13

 usually

1.10

Activations Density 2.498%