INDEX

Explanations

phrases related to sharing or protecting personal data

oai_token-act-pair · gpt-3.5-turbo

references to personal data and information privacy

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ajor

-0.80

STRUCT

-0.73

 McDonnell

-0.73

unct

-0.72

ModLoader

-0.71

sym

-0.68

yrinth

-0.68

 Instruction

-0.67

ithing

-0.66

addle

-0.66

POSITIVE LOGITS

 encrypted

1.00

 collected

0.95

 breaches

0.95

 passwords

0.92

 privacy

0.92

 inappropriately

0.91

 unlawfully

0.91

 trove

0.91

 encryption

0.88

 analytics

0.87

Activations Density 0.092%

phrases related to sharing or protecting personal data

references to personal data and information privacy

No Comments

No Known Activations