INDEX

Explanations

numbers and characters

The neuron activates on numerical tokens, i.e. isolated digits or numbers 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

",

-1.91

");

-1.87

釭

-1.83

主に

-1.77

 they

-1.77

 pesta

-1.74

ൎ

-1.74

鲮

-1.73

 tragé

-1.72

⁇

-1.72

POSITIVE LOGITS

4.53

and

2.27

</strong>

2.17

</sub>

2.05

却不

1.89

是没有

1.88

<em>

1.88

 respostas

1.88

⁛

1.83

1.80

Activations Density 0.028%