INDEX

Explanations

recent events or developments

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

-0.08

deaux

-0.08

Yam

-0.08

-redux

-0.08

&R

-0.08

ä½ľçĶ¨

-0.08

esi

-0.08

ses

-0.08

POSITIVE LOGITS

 events

0.23

 developments

0.19

 news

0.18

 happen

0.17

äºĭä»¶

0.17

events

0.16

 aconte

0.16

 Events

0.15

 event

0.15

-events

0.15

Activations Density 0.082%