INDEX

Explanations

primary

The neuron is tuned to detect the word “primary,” especially when it appears as an adjective (e.g. primary care, primary school, primary source).

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

த்தி

0.61

 Carrasco

0.59

 ayrıca

0.58

وو

0.57

шт

0.57

 продов

0.56

spike

0.55

 हानि

0.55

وینت

0.55

 ชี

0.55

POSITIVE LOGITS

 Primary

1.10

 primary

1.06

Primary

0.98

 первич

0.91

primary

0.88

 PRIMARY

0.88

 प्राथमिक

0.86

 प्राइमरी

0.85

 primaries

0.84

 Secondary

0.83

Activations Density 0.050%