INDEX

Explanations

architecturally significant

New Auto-Interp

Top Features by Cosine Similarity

Configuration

Prompts (Dashboard)

10,000 prompts, 128 tokens each

Dataset (Dashboard)

lmsys/lmsys-chat-1m

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 bund

-0.10

Hak

-0.09

rug

-0.09

 Gutenberg

-0.09

VML

-0.09

pig

-0.09

agem

-0.09

ama

-0.09

asse

-0.09

 Desk

-0.09

POSITIVE LOGITS

 architecture

0.17

itecture

0.17

Architecture

0.16

 ÙħØ¹ÙħØ§Ø±ÛĮ

0.16

architecture

0.16

 Architecture

0.16

 Ð°ÑĢÑħ

0.15

arch

0.15

 architectural

0.15

 patterns

0.14

Activations Density 0.069%