INDEX

Explanations

references to privacy-related terms

oai_token-act-pair · gpt-3.5-turbo

references to privacy

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

Production

-0.79

×Ļ×

-0.74

hyde

-0.70

WAYS

-0.70

xual

-0.70

shi

-0.64

INK

-0.64

Job

-0.63

à¤

-0.63

ACTED

-0.63

POSITIVE LOGITS

 privacy

1.12

 protections

0.83

 liberties

0.80

 safeguards

0.79

 suits

0.79

 rights

0.79

 anonymity

0.77

parency

0.75

 confidentiality

0.71

 Liberties

0.70

Activations Density 0.013%

references to privacy-related terms

references to privacy

No Comments

No Known Activations