INDEX

Explanations

phrases related to discrimination and prejudice, particularly focusing on sexism

oai_token-act-pair · gpt-3.5-turbo

terms related to sexism and misogyny

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Trust

-0.76

leaf

-0.73

Package

-0.71

hyde

-0.71

mental

-0.69

ernels

-0.68

uilding

-0.68

VIS

-0.68

ving

-0.67

NAS

-0.67

POSITIVE LOGITS

 sexist

1.03

 misogyn

0.89

 slurs

0.87

 jokes

0.81

 stereotypes

0.79

 stereotyp

0.78

 banter

0.76

 Equality

0.74

 feminists

0.73

 sexism

0.73

Activations Density 0.018%

phrases related to discrimination and prejudice, particularly focusing on sexism

terms related to sexism and misogyny

No Comments

No Known Activations