INDEX

Explanations

translated something to

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

ç¸¾

-0.10

ä¹Ĥ

-0.10

ï»¿#

-0.09

å¼ĭ

-0.09

Tib

-0.09

<|begin_of_text|>

-0.09

 è©ķä¾¡

-0.09

 çĶ³åįļ

-0.09

 Pilot

-0.09

:::::::::::::::

-0.09

POSITIVE LOGITS

ãĥ¼

0.13

Sen

0.12

Sen

0.11

ãĢĢ

0.11

ï½

0.11

Eig

0.10

okane

0.10

andom

0.10

Nik

0.10

 Western

0.09

Activations Density 0.007%