INDEX

Explanations

function calls or delimiters

The neuron activates primarily on numeric tokens, i.e. instances of numbers (digits) in the text.

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

 arşivlendi

-1.19

 (!__

-1.13

 عاما

-1.12

 violación

-1.05

quê

-1.03

results

-0.99

क्षक

-0.96

 ماز

-0.94

 meest

-0.93

ldquo

-0.91

POSITIVE LOGITS

2.86

2.13

}(

1.34

$(

1.21

()):

1.08

waniem

1.08

>(</

1.07

 there

1.05

()),

1.05

<>(

1.04

Activations Density 0.035%