INDEX

Explanations

diplomatic success, situation, or service

The neuron primarily activates on words and word‐pieces related to “diplomatic” or “diplomacy.”

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

It

-2.78

–

-2.64

 lament

-2.63

࿚

-2.63

がない

-2.61

Its

-2.53

食物

-2.48

 Coupled

-2.45

 surpass

-2.44

 replete

-2.34

POSITIVE LOGITS

 parachoque

2.80

ﱠ

2.69



2.59

缫

2.58

 駅前

2.50

寶

2.50

龌

2.48

蚍

2.48

ſſel

2.47

Ꮸ

2.47

Activations Density 0.001%