INDEX

Explanations

code block markdown

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

 nuclear

-0.11

 Nuclear

-0.10

Sil

-0.10

 Stokes

-0.09

uclear

-0.09

sen

-0.09

sen

-0.09

Opens

-0.08

 Booth

-0.08

Charset

-0.08

POSITIVE LOGITS

 block

0.18

 code

0.14

 fenced

0.14

 CODE

0.14

(code

0.14

 fences

0.14

.code

0.14

 Code

0.13

.setCode

0.13

 blocks

0.13

Activations Density 0.119%