INDEX

Explanations

measurements and service details

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

 timelines

-0.10

otel

-0.09

ddit

-0.09

orp

-0.09

bu

-0.09

$__

-0.09

groupon

-0.08

 Marr

-0.08

Intialized

-0.08

odia

-0.08

POSITIVE LOGITS

 frequency

0.34

 times

0.25

frequency

0.24

 Frequency

0.23

 length

0.21

 amount

0.20

 frequencies

0.20

Frequency

0.19

 duration

0.19

-frequency

0.18

Activations Density 0.153%