INDEX

Explanations

matrix class command folder

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

įng

-0.10

 -*-č\n

-0.10

hlas

-0.09

łéĻ¤

-0.09

idl

-0.08

 inherits

-0.08

edith

-0.08

imbus

-0.08

¨ë¶Ģ

-0.08

OLVE

-0.08

POSITIVE LOGITS

reen

0.09

«

0.08

createForm

0.08

 Walton

0.08

 surg

0.08

quila

0.08

Leap

0.08

inta

0.08

rep

0.07

/_

0.07

Activations Density 0.051%