INDEX

Explanations

customizable options

The neuron fires on words that describe providing or ensuring a supportive, protective, or enriching environment (e.g. managing, making, protected, learning, experience, provide).

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

ewe

0.56

ванные

0.55

ajú

0.55

зю

0.54

੦

0.54

awca

0.54

헉

0.54

再來

0.53

 বিভিন্ন

0.52

ajuan

0.52

POSITIVE LOGITS

 samtidigt

2.55

 samtidig

2.50

的同时

2.36

 jednocześnie

2.36

 WHILE

2.33

 simultaneously

2.17

 while

2.16

同时

2.14

 sekaligus

2.13

 zároveň

2.11

Activations Density 0.545%