INDEX

Explanations

controversial or conflicting statements regarding health and medicine, such as about vaccination safety, processed meats, and the impact of chemicals

oai_token-act-pair · gpt-3.5-turbo

terms and phrases related to health risks and medical claims

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

igsaw

-0.82

 Horizons

-0.76

 Bezos

-0.75

 Schn

-0.75

owski

-0.70

 Franchise

-0.70

utenberg

-0.69

ahime

-0.69

psons

-0.68

izons

-0.68

POSITIVE LOGITS

 harmful

1.66

 carcin

1.65

 addictive

1.54

 toxins

1.33

 unsafe

1.33

 neurot

1.33

 harmless

1.32

 unhealthy

1.30

 safer

1.30

 detrimental

1.29

Activations Density 0.342%

controversial or conflicting statements regarding health and medicine, such as about vaccination safety, processed meats, and the impact of chemicals

terms and phrases related to health risks and medical claims

No Comments

No Known Activations