INDEX

Explanations

developed a

The neuron strongly activates on the verb “develop” (in all its morphological forms, e.g. develop, developed, developing).

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

𓋼

-1.77

慙

-1.75

想着

-1.59

without

-1.58



-1.57

埚

-1.55

跩

-1.55

 تحقی

-1.54

 predominantly

-1.53

偷偷

-1.52

POSITIVE LOGITS

During

1.63

</h2>

1.61

 weirdly

1.57

’

1.56

，

1.47

 Usually

1.45

?”

1.43

 excessively

1.43

After

1.42

自己的

1.42

Activations Density 0.052%