INDEX

Explanations

including various types

The neuron activates on numeric expressions (digits and numbers) 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

at

-1.26

or

-1.25

 didn

-1.15

 when

-1.10

 should

-1.08

 lainnya

-1.05

 that

-1.05

 other

-1.02

 Only

-1.02

 ONLY

-1.01

POSITIVE LOGITS

 such

1.65

 vélos

1.34

常见的

1.26

 ranging

1.24

 такие

1.22

 таких

1.20

among

1.17

 myriad

1.14

ături

1.14

常見

1.13

Activations Density 0.097%