INDEX

Explanations

teams' match results

The neuron is detecting mentions of sports team names.

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

与其他

0.63

 sports

0.60

 اللاعب

0.49

 continuar

0.49

スポーツ

0.49

ظر

0.48

 continu

0.48

球队

0.48

 Sports

0.48

されていない

0.48

POSITIVE LOGITS

whom

0.70

 endet

0.58

 whom

0.57

𝘷

0.57

СТЕ

0.56

decoded

0.55

 foiled

0.55

𝙛

0.55

decrypt

0.55

 ফোনে

0.54

Activations Density 0.017%