INDEX

Explanations

small entities with contrast

The neuron is triggered by mentions of smallness—that is, words indicating a small or tiny size or scale (e.g. “small,” “tiny,” etc.).

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

 wezen

-1.00

 wraz

-0.98

 kolei

-0.94

 through

-0.91

ípio

-0.89

cún

-0.88

CHREIB

-0.88

Ϝ

-0.88

;'>

-0.87

 նաև

-0.87

POSITIVE LOGITS

but

2.28

 लेकिन

1.31

 compared

1.29

 nhưng

1.27

yet

1.24

(<

1.23

compared

1.23

its

1.07

 pero

1.05

但却

1.05

Activations Density 0.046%