INDEX

Explanations

end of / recipient of

This neuron detects idiomatic “end” phrases that denote a position or outcome (e.g. “receiving end,” “losing end,” “end of the stick”).

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

-2.44

-2.13

 other

-1.83

其他的

-1.63

But

-1.62

 recent

-1.49

その他の

-1.47

是指

-1.45

-1.41

 June

-1.40

POSITIVE LOGITS

 этих

2.00

 ヌード

1.71

 撮っ

1.71

梣

1.69

Смотрите

1.63

Выбор

1.61

Эти

1.59

viembre

1.55

FUENTES

1.55

𓐍

1.54

Activations Density 0.020%