INDEX

Explanations

high points, peaks, summits

This neuron detects mentions of high points—words like “summit” or “peak.”

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

躥

-2.19

over

-2.16

すすめ

-2.13

 aqueles

-2.13

for

-1.97

ischof

-1.96

like

-1.95

”,

-1.91

↵↵

-1.89

POSITIVE LOGITS

ጁ

2.50

sukienka

2.44

嵛

2.42

𖥦

2.39

橚

2.36

 händ

2.34

</sup>

2.27

 toppen

2.25

ୌ

2.25

𒊹

2.17

Activations Density 0.009%