INDEX

Explanations

now, once, still, usually, often followed by descriptor

The neuron is triggered by numeric tokens (especially those containing digits and hyphens).

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

を

-1.84

が

-1.55

aneity

-1.55

それにしても

-1.51

に

-1.49

itisation

-1.48

ternut

-1.45

natured

-1.40

などの

-1.39

zied

-1.38

POSITIVE LOGITS

ようです

1.45

為に

1.45

際に

1.27

際の

1.14

#%%

1.13

際には

1.06

たびに

1.06

Similarly

1.06

/*!

1.04

 equally

1.03

Activations Density 0.097%