INDEX

Explanations

phrases or terms used to describe or categorize people by socioeconomic status or behavior

oai_token-act-pair · gpt-3.5-turbo

references to social labels or derogatory terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Solitaire

-0.83

rift

-0.81

tains

-0.78

 inhib

-0.71

tainment

-0.69

lessly

-0.68

 Roller

-0.63

 Tycoon

-0.63

 promo

-0.63

rob

-0.62

POSITIVE LOGITS

glomer

0.76

anges

0.71

ural

0.71

 è£ıè¦ļéĨĴ

0.69

 Brach

0.69

ethy

0.69

avorite

0.67

afort

0.67

BuyableInstoreAndOnline

0.67

rontal

0.66

Activations Density 0.000%

phrases or terms used to describe or categorize people by socioeconomic status or behavior

references to social labels or derogatory terms

No Comments

No Known Activations