INDEX

Explanations

I

This neuron detects first-person self-references—especially the pronoun “I” (and occasionally “we”) indicating the speaker’s own statements.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

0.39

↵

0.34

0.32

by

0.29

<0x0D>

0.29

0.27

::

0.27

POSITIVE LOGITS

 Didn

0.31

 എന്റെ

0.31

 জানি

0.31

 pensé

0.31

 persönlich

0.30

 বললাম

0.30

看到了

0.30

ਨ੍ਹਾਂ

0.30

 দেখছি

0.30

 너무

0.29

Activations Density 1.224%