INDEX

Explanations

terms related to different demographic characteristics or categories such as race, ethnicity, nationality, religion, and disabilities

oai_token-act-pair · gpt-3.5-turbo

terms related to social identity and discrimination categories

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

writers

-0.79

 Raphael

-0.72

 Peng

-0.67

sers

-0.66

 Canaver

-0.65

 Kers

-0.64

Byr

-0.64

said

-0.64

HEL

-0.63

 Dean

-0.62

POSITIVE LOGITS

 ethnicity

1.69

 nationality

1.59

 gender

1.54

Gender

1.39

 ethnic

1.37

 Gender

1.33

 creed

1.28

 sexuality

1.27

gender

1.26

 Ethnic

1.25

Activations Density 0.171%

terms related to different demographic characteristics or categories such as race, ethnicity, nationality, religion, and disabilities

terms related to social identity and discrimination categories

No Comments

No Known Activations