INDEX

Explanations

track objects

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

 layer

-0.09

_detector

-0.09

 Equal

-0.09

 standing

-0.09

ctp

-0.09

 diagn

-0.09

gas

-0.08

 Bran

-0.08

Visualization

-0.08

 layers

-0.08

POSITIVE LOGITS

 tracking

0.30

 Tracking

0.28

Tracking

0.25

 track

0.24

tracking

0.23

 Track

0.23

 tracked

0.22

Track

0.22

 tracker

0.21

 trackers

0.21

Activations Density 0.044%