INDEX

Explanations

specific mentions of the concept of privacy

oai_token-act-pair · gpt-3.5-turbo

words related to privacy or privileged information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

calling

-0.74

ORN

-0.73

grass

-0.70

IELD

-0.68

LER

-0.68

WAYS

-0.68

 Shake

-0.67

 Ducks

-0.65

 Pixie

-0.65

 Centauri

-0.64

POSITIVE LOGITS

acies

1.36

ilege

1.28

ileged

1.27

ately

1.22

ilage

1.07

atism

0.97

aband

0.96

acy

0.96

tesy

0.93

iless

0.88

Activations Density 0.036%

specific mentions of the concept of privacy

words related to privacy or privileged information

No Comments

No Known Activations