INDEX

Explanations

terms related to discrimination based on characteristics such as race, ethnicity, nationality, religion, and disability

oai_token-act-pair · gpt-3.5-turbo

references to social categories related to identity and discrimination

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ahead

-0.75

Byr

-0.75

 Drawn

-0.68

nels

-0.68

ateurs

-0.67

mr

-0.67

rats

-0.66

reb

-0.65

irect

-0.65

POSITIVE LOGITS

 ethnicity

1.86

 nationality

1.75

 gender

1.60

 ethnic

1.49

 creed

1.43

 Gender

1.40

 Ethnic

1.37

 socioeconomic

1.35

Gender

1.30

 sexuality

1.30

Activations Density 0.120%

terms related to discrimination based on characteristics such as race, ethnicity, nationality, religion, and disability

references to social categories related to identity and discrimination

No Comments

No Known Activations