INDEX

Explanations

Gal, garlic, Gast, Galois

This neuron detects capitalized proper nouns (names of people, places, or organizations).

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

biv

0.59

 suffixes

0.58

 tapi

0.57

 hintergrund

0.54

 vall

0.54

氾

0.52

渽

0.52

رفت

0.52

 biết

0.51

sa

0.51

POSITIVE LOGITS

 га

0.94

 Га

0.81

untlet

0.76

GA

0.69

Gab

0.63

Га

0.60

Gau

0.60

riele

0.60

Gah

0.59

Ga

0.59

Activations Density 0.256%