INDEX

Explanations

a survey / colon This

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

 caller

-0.09

ibold

-0.09

ihil

-0.09

 studies

-0.09

udies

-0.09

ettel

-0.09

 Caller

-0.08

eum

-0.08

 polled

-0.08

EDGE

-0.08

POSITIVE LOGITS

 snow

0.17

Lik

0.17

 convenience

0.16

lik

0.15

 Convenience

0.14

snow

0.13

 purpos

0.13

 Snow

0.13

Lik

0.12

 triang

0.12

Activations Density 0.036%