INDEX

Explanations

coded or suggestive language that communicates

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

zew

-0.11

 unknown

-0.10

 Holt

-0.09

enth

-0.08

 éĸ

-0.08

unknown

-0.08

 Unknown

-0.08

 mention

-0.08

 CHUNK

-0.08

ï¾

-0.08

POSITIVE LOGITS

 layers

0.20

 meanings

0.20

 layer

0.20

 meaning

0.19

 Layer

0.18

 hidden

0.18

 alleg

0.17

 deeper

0.17

layers

0.17

 Layers

0.17

Activations Density 0.150%