INDEX

Explanations

variable declarations ending in semicolon

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

 import

-0.12

vic

-0.11

vic

-0.11

-0.10

 Micha

-0.10

Vic

-0.09

bew

-0.09

 Mang

-0.09

if

-0.09

POSITIVE LOGITS

;\n

0.17

;\n\n

0.17

0.13

jspx

0.11

;</

0.11

;\n\n\n

0.10

;//

0.10

;\n

0.10

.*;\n\n

0.10

 {};\n

0.10

Activations Density 0.052%