INDEX

Explanations

concepts followed by confirming verbs

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

azar

-0.09

ione

-0.09

age

-0.09

PTY

-0.09

neys

-0.09

oric

-0.09

ripe

-0.09

 provided

-0.09

POSITIVE LOGITS

 earned

0.17

earned

0.13

 earn

0.13

 hasn

0.12

 extends

0.12

 earning

0.11

 extend

0.11

led

0.11

 dates

0.11

 apparent

0.11

Activations Density 0.009%