INDEX

Explanations

mentions of the name "Hom" with varying levels of importance

oai_token-act-pair · gpt-3.5-turbo

references to homosexuality and related topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

DAQ

-0.77

 refresh

-0.75

acca

-0.70

OWER

-0.69

daq

-0.69

 Solitaire

-0.68

cc

-0.67

ragon

-0.66

BE

-0.66

ofi

-0.65

POSITIVE LOGITS

Hom

3.80

Hom

3.15

hom

3.10

hom

1.99

 homophobia

1.40

 homosexuals

1.35

 homophobic

1.29

 homosexual

1.28

 homosexuality

1.27

Gay

1.25

Activations Density 0.025%

mentions of the name "Hom" with varying levels of importance

references to homosexuality and related topics

No Comments

No Known Activations