INDEX

Explanations

E.g

This neuron detects the token “E.” (an uppercase E followed by a period), as used in examples or citation-style initials.

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

-2.56

</b>

-2.53

-2.23

ي

-2.13

},

-2.11

termilk

-2.09

</td>

-2.08

،

-1.95

```

-1.93

");

-1.91

POSITIVE LOGITS

 This

2.50

 này

2.30

 viaja

2.22

isn

2.22

 What

2.14

 alberga

2.11

 mengapa

2.03

ཫ

2.02

鹮

1.98

 actúa

1.95

Activations Density 0.026%