INDEX

Explanations

nn.Linear layer dimensions

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

 strand

-0.10

 Peripheral

-0.10

raman

-0.09

 Strand

-0.09

atatype

-0.09

 peripheral

-0.09

 Criminal

-0.08

 Colomb

-0.08

ns

-0.08

POSITIVE LOGITS

0.16

0.15

0.14

0.11

vanced

0.10

rous

0.10

éĴ®

0.09

 hidden

0.09

ActionCreators

0.09

ÐĶÐĲ

0.09

Activations Density 0.013%