INDEX

Explanations

phrases related to security

oai_token-act-pair · gpt-3.5-turbo

references to security measures and concerns

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

poons

-0.80

etus

-0.78

 Johann

-0.75

aho

-0.73

album

-0.70

inen

-0.69

udder

-0.69

 Staten

-0.68

igne

-0.67

aning

-0.67

POSITIVE LOGITS

 security

3.69

security

3.03

Security

2.90

 Security

2.88

 cybersecurity

1.95

 safety

1.66

 privacy

1.65

 counterterrorism

1.50

 insecurity

1.47

 surveillance

1.39

Activations Density 0.025%

phrases related to security

references to security measures and concerns

No Comments

No Known Activations