INDEX

Explanations

self-referential statements about the assistant’s identity and capabilities, especially first-person mentions about being a model and how it operates.

New Auto-Interp

Configuration

Prompts (Dashboard)

238,145 prompts, 512 tokens each

Dataset (Dashboard)

lmsys + oasst1

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 требуют

0.66

 ক্রয়

0.65

 incorrect

0.64

 worsen

0.62

 предлагают

0.61

要注意

0.61

 khuyến

0.61

 loạt

0.60

 dubious

0.60

 ማስ

0.58

POSITIVE LOGITS

 myself

1.37

 Myself

1.30

 আমার

1.23

 আমাকে

1.21

 আমি

1.16

我的

1.13

 meinem

1.11

 我的

1.10

我是

1.09

 제가

1.09

Activations Density 0.023%