INDEX

Explanations

past tense verbs followed by common words

The main thing this neuron does is pick out unusually long or multisyllabic words.

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

だけではなく

-1.38

に行く

-1.34

 reveló

-1.27

 zwanzig

-1.27

 supuestamente

-1.26

に行った

-1.23

 OGSÅ

-1.23

 dieser

-1.22

了很多

-1.16

 trabaja

-1.15

POSITIVE LOGITS

まずは

1.44

幼稚園

1.15

写真は

1.13

(*´

1.12

 domés

1.10

や

1.10

 tiro

1.09

 various

1.06

最近は

1.05

日時

1.05

Activations Density 0.040%