INDEX

Explanations

consider structure

The neuron strongly activates on personal pronouns (I, we, you, my, our) and nearby common verbs, identifying segments where the author speaks in the first or second person.

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

 hotel

0.86

 hotels

0.76

 tourist

0.75

 touristic

0.71

 Hotel

0.71

 Hotels

0.71

 guest

0.70

hotel

0.70

 guests

0.70

 tourism

0.69

POSITIVE LOGITS

 Beob

0.68

 Tät

0.64

構築

0.61

食べる

0.61

 Bereiche

0.60

 กิน

0.59

 ベッド

0.59

を活用

0.59

を用いた

0.58

 উপভোগ

0.58

Activations Density 0.106%