INDEX

Explanations

words related to confidentiality and sensitive information

oai_token-act-pair · gpt-3.5-turbo

references to sensitive information and issues surrounding it

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

FIN

-0.85

AUT

-0.79

 Wolver

-0.75

 Truck

-0.74

swick

-0.71

amaz

-0.70

LOAD

-0.70

 Helsinki

-0.69

UNCH

-0.69

SN

-0.68

POSITIVE LOGITS

 sensitive

1.25

ivities

1.00

mble

0.95

 sensit

0.87

ensitive

0.83

 sensitivity

0.81

ively

0.81

sensitive

0.79

itized

0.79

 sensibilities

0.78

Activations Density 0.011%

words related to confidentiality and sensitive information

references to sensitive information and issues surrounding it

No Comments

No Known Activations