INDEX

Explanations

more and more common

The neuron detects the repeated intensifier pattern “more and more.”

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

But

-2.09

of

-1.94

 Suiza

-1.92

堪能

-1.83

 akhirnya

-1.78

 pero

-1.70

々木

-1.70

覗

-1.63

更に

-1.60

but

-1.59

POSITIVE LOGITS

2.09

ly

1.91

んでいる

1.69

the

1.63

 стоковая

1.60

 feiert

1.60

platten

1.59

gefügt

1.57

drück

1.56

不

1.52

Activations Density 0.020%