INDEX

Explanations

microphone or speaker

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

åĤ¬

-0.10

dna

-0.09

IRR

-0.09

dna

-0.09

 Transparent

-0.09

,ï½¤

-0.09

 quang

-0.09

rieg

-0.09

DVD

-0.09

Fuse

-0.09

POSITIVE LOGITS

mic

0.30

 microphone

0.28

Mic

0.26

mic

0.26

 Micro

0.23

Mic

0.23

éº¦

0.22

Micro

0.21

micro

0.20

 micro

0.20

Activations Density 0.063%