INDEX

Explanations

working with others on tasks

The neuron fires on occurrences of the verb “work” (and its inflected forms like “worked,” “working,” “works”) indicating mentions of someone working or collaborating.

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

ında

1.46

ının

1.35

ladı

1.30

wechsel

1.29

时候

1.27

ंनी

1.27

कने

1.27

qu

1.26

ce

1.26

bruk

1.22

POSITIVE LOGITS

 tirelessly

1.48

াকিস্ত

1.31

 étroite

1.30

Kết

1.24

НИ

1.14

 kinks

1.14

 diligence

1.13

ກັບ

1.13

 isolate

1.11

 DISABLE

1.11

Activations Density 0.177%