INDEX

Explanations

alca, alco, alkal prefixes

The neuron activates on tokens containing the letter sequence “alk” (e.g. Alka-, Alc-, alkoxy, Alkacon).

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

setViewName

-0.80

 اسم

-0.73

öszi

-0.73

虑

-0.72

Bez

-0.70

 Ropa

-0.69

REEK

-0.69

FILM

-0.69

Concepto

-0.68

FMC

-0.68

POSITIVE LOGITS

Alc

0.76

alk

0.72

 Howe

0.71

chel

0.70

Losing

0.69

 alkoh

0.69

vssd

0.68

jaz

0.67

'*

0.66

 Voyez

0.66

Activations Density 0.053%