INDEX

Explanations

sizes and measurements

The neuron detects numeric measurement values (e.g., dimensions, sizes, and decimal quantities) in the text.

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

を入れる

-1.18

-1.15

をチェック

-1.13

rée

-1.11

 повсе

-1.10

 houver

-1.09

以外にも

-1.09

 annihilated

-1.07

простра

-1.07

 Herstellung

-1.06

POSITIVE LOGITS

 berangkat

1.30

ólico

1.25

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.24

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.22

getLocal

1.20

 anvä

1.20

日起

1.20

alisme

1.18

 bergerak

1.17

 libras

1.17

Activations Density 0.030%