INDEX

Explanations

wave, waves, wavy

The neuron fires strongly on any token containing “wave” (e.g. Wave, waveform, waves, etc.).

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

.”

-2.41

出来ました

-2.25

 شیپور

-2.25

.";

-2.23

日曜日

-2.16

逦

-2.08

 ragaz

-2.06

 古风

-1.96

在這

-1.95

';

-1.94

POSITIVE LOGITS

2.55

 literalmente

2.00

 originalmente

1.88

 waves

1.88

 Wave

1.88

 industrious

1.82

ll

1.80

 вопрос

1.80

Activations Density 0.022%