INDEX

Explanations

special characters or bytes

The neuron strongly responds to abstract content‐laden terms (mostly academic or formal nouns and related function words) characteristic of scholarly or technical prose.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

<0x0D>

0.26

0.25

0.23

SELF

0.23

pH

0.22

CATEGORY

0.22

 服务

0.22

aceous

0.22

POSITIVE LOGITS

트를

0.25

というのが

0.25

 যদি

0.25

<unused1771>

0.24

<unused1918>

0.24

<unused410>

0.24

<unused1938>

0.24

ót

0.24

 सीमित

0.24

ത്തിനായി

0.24

Activations Density 0.018%