INDEX

Explanations

levels and degrees

This neuron detects words that denote hierarchical or graded levels (e.g. tiered, graded, hierarchical).

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

 Those

-1.65

'';

-1.63

 That

-1.59

轫

-1.58

seded

-1.58

 ハロウィン

-1.57

istungen

-1.52

 Jurassic

-1.51

 Often

-1.49

 Thanks

-1.48

POSITIVE LOGITS

."

1.86

different

1.70

 diferentes

1.69

</u>

1.66

how

1.60

 altamente

1.53

 distintos

1.51

 różnych

1.48

.'

1.46

1.45

Activations Density 0.045%