INDEX

Explanations

words related to negative social perceptions such as stigma, shame, and discrimination

oai_token-act-pair · gpt-3.5-turbo

terms related to stigma and stigmatization

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

apeake

-0.78

tes

-0.72

ramid

-0.72

ebus

-0.71

hire

-0.71

ropolitan

-0.69

annis

-0.67

semble

-0.67

 Eucl

-0.66

irgin

-0.65

POSITIVE LOGITS

 stigma

1.15

 stigmat

0.87

imaru

0.78

ously

0.77

 prejudice

0.69

 Shame

0.69

ovan

0.67

uously

0.67

 attached

0.67

endered

0.65

Activations Density 0.013%

words related to negative social perceptions such as stigma, shame, and discrimination

terms related to stigma and stigmatization

No Comments

No Known Activations