INDEX

Explanations

terms related to the LGBTQ+ community, specifically lesbian individuals

oai_token-act-pair · gpt-3.5-turbo

references to lesbian and LGBTQ+ themes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ULE

-0.82

æĸ¹

-0.81

EY

-0.80

frames

-0.74

rex

-0.72

Anim

-0.71

schild

-0.71

AH

-0.70

ROR

-0.70

lamm

-0.69

POSITIVE LOGITS

 lesbian

0.99

 couples

0.95

ism

0.90

 Lesbian

0.87

sex

0.87

 bisexual

0.87

 emancipation

0.86

ization

0.86

 separat

0.86

 heterosexual

0.81

Activations Density 0.009%

terms related to the LGBTQ+ community, specifically lesbian individuals

references to lesbian and LGBTQ+ themes

No Comments

No Known Activations