INDEX

Explanations

sarcastic, bizarre, hilarious

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

 decent

-0.12

-0.10

linger

-0.10

ouch

-0.09

 politics

-0.09

æľīçĤ¹

-0.09

 wires

-0.09

 Signature

-0.09

POSITIVE LOGITS

riv

0.14

 swell

0.12

 subtle

0.12

 sarc

0.11

 totally

0.11

oh

0.11

riv

0.11

 absolutely

0.10

 whopping

0.10

(/

0.09

Activations Density 0.068%