INDEX

Explanations

job displacement

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

ï¾ŀ

-0.10

 thÃ´i

-0.09

lil

-0.09

 Civil

-0.09

DidAppear

-0.09

ORMAT

-0.09

 loos

-0.09

 slap

-0.08

 decentral

-0.08

ormal

-0.08

POSITIVE LOGITS

job

0.19

job

0.18

 workers

0.17

 worker

0.16

 displacement

0.16

Job

0.14

 jobs

0.13

 displaced

0.13

 Transition

0.13

/job

0.13

Activations Density 0.026%