INDEX

Explanations

acting together

The neuron strongly fires on the word “together” (i.e. expressions of joint or collective action).

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

見えます

-1.48

-1.45

 duplicate

-1.44

 successor

-1.39

 nuevamente

-1.38

 extraño

-1.38

 real

-1.35

 glimpse

-1.33

意味着

-1.33

 zwischen

-1.30

POSITIVE LOGITS

 with

1.66

ⓒ

1.63

iocene

1.54

."));

1.53

czeń

1.46

тивный

1.44

极为

1.43

;");

1.40

эль

1.40

AndGet

1.39

Activations Density 0.025%