INDEX

Explanations

unclear

The neuron specializes in detecting words and phrases that express scientific uncertainty or unknown/unclarified status (e.g., “unclear,” “unknown”).

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

in

-2.59

on

-2.58

the

-2.33

for

-2.22

 between

-2.05

with

-1.84

-1.76

 through

-1.75

 thru

-1.70

via

-1.60

POSITIVE LOGITS

 Minggu

1.74

Ṇ

1.67

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.66

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.64

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.61

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.61

º

1.60

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.60

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.59

 jadx

1.57

Activations Density 0.041%