INDEX

Explanations

going to choose

The neuron strongly activates on first- and second-person pronouns (especially “I” and “you”).

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

不断

0.69

 periodically

0.66

ことがある

0.64

不斷

0.63

がたくさん

0.62

 регулярно

0.61

!\!\

0.60

知らせ

0.59

 ständig

0.59

があると

0.59

POSITIVE LOGITS

 opting

1.86

 chose

1.81

 opted

1.73

 choosing

1.73

选择了

1.72

 choose

1.62

选择

1.60

 choisir

1.55

 선택

1.55

 optar

1.54

Activations Density 0.844%