INDEX

Explanations

The neuron is firing on long sequences of whitespace or punctuation used as document formatting separators.

parentheticals, lists, and questions

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

 lantaran

-1.13

our

-1.12

 його

-1.11

-1.03

Following

-0.96

 most

-0.93

ʽ

-0.91

 traditionelle

-0.90

🔟

-0.88

stanford

-0.88

POSITIVE LOGITS

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.23

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.21

↵↵↵↵↵↵↵↵↵↵↵↵↵

1.20

↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.19

↵↵↵↵↵↵↵↵↵↵

1.16

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.16

nehå

1.14

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.13

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.13

↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵

1.09

Activations Density 0.000%