INDEX

Explanations

concavity, compactness

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

 Lionel

-0.10

 Anatomy

-0.09

Dur

-0.09

 Cruiser

-0.09

cot

-0.08

ButtonDown

-0.08

 Shelf

-0.08

 reve

-0.08

 Sands

-0.08

icopter

-0.08

POSITIVE LOGITS

 shape

0.12

Lob

0.12

 duct

0.12

 perimeter

0.11

Ecc

0.11

.texture

0.11

 feature

0.11

 eccentric

0.11

 area

0.10

ipple

0.10

Activations Density 0.024%