INDEX

Explanations

abstract concepts and topics

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

vero

-0.09

 forth

-0.09

 Ð¾ÑĤÐ»Ð¸

-0.09

 predic

-0.09

clare

-0.08

imon

-0.08

 Lazar

-0.08

afil

-0.08

Alv

-0.08

POSITIVE LOGITS

 matter

0.25

 topic

0.24

matter

0.19

 subject

0.18

 topics

0.17

topic

0.16

 Matter

0.16

 matters

0.16

 SUBJECT

0.15

subject

0.15

Activations Density 0.067%