INDEX

Explanations

suffer from

The neuron fires on occurrences of the verb “suffer” (and its inflected forms like “suffers” or “suffered”).

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

the

-2.02

 xícara

-1.90

Our

-1.85

😃

-1.80

你来

-1.72

蒟

-1.72

 撮っ

-1.71

が上が

-1.70

 наша

-1.68

が取

-1.67

POSITIVE LOGITS

৫

1.77

1.75

”,

1.73

蘼

1.72

 menetapkan

1.70

自己的

1.69

ts

1.69

 angegebenen

1.64

</h2>

1.60

or

1.60

Activations Density 0.013%