INDEX

Explanations

via a, RNA, be joined

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

 compound

-0.13

 Compound

-0.13

Compound

-0.13

 genomic

-0.11

 prote

-0.11

compound

-0.10

 Prote

-0.10

 protein

-0.09

 proteins

-0.09

DDS

-0.09

POSITIVE LOGITS

 primer

0.21

 prim

0.20

primer

0.17

Ta

0.17

 Primer

0.16

 nick

0.15

 templates

0.14

 reactions

0.14

Anne

0.14

anne

0.14

Activations Density 0.022%