INDEX

Explanations

references to sins, especially related to ethical or moral wrongdoing

oai_token-act-pair · gpt-3.5-turbo

references to the concept of sin

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

quickShipAvailable

-0.76

DragonMagazine

-0.75

æ©Ł

-0.72

PsyNetMessage

-0.70

awei

-0.68

poons

-0.68

berman

-0.67

IFE

-0.65

 Aerospace

-0.64

renches

-0.63

POSITIVE LOGITS

ned

1.15

ning

0.98

fulness

0.92

fully

0.91

ews

0.89

uous

0.83

ners

0.81

acious

0.77

 Sins

0.76

ters

0.75

Activations Density 0.022%

references to sins, especially related to ethical or moral wrongdoing

references to the concept of sin

No Comments

No Known Activations