INDEX

Explanations

Supporting Information files

This neuron detects references to supplemental materials—like figures, legends, tables, or “Supporting Information”—along with their associated numbering.

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

 which

-1.58

 with

-1.44

 whose

-1.27

 祝福

-1.14

to

-1.14

想象

-1.07

 hampir

-1.07

 without

-1.05

↵

-1.04

 podrá

-1.03

POSITIVE LOGITS

の説明

1.13

[*]

1.09

ukų

1.06

 trenes

1.06

เพิ่มเติม

1.06

 décès

1.05

 boire

1.03

 comptes

1.03

 laiss

1.03

 choroby

1.02

Activations Density 0.005%