INDEX

Explanations

phrases indicating eligibility or ineligibility for advertising

oai_token-act-pair · gpt-3.5-turbo

references to eligibility for advertising

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

hess

-0.84

afety

-0.80

Git

-0.79

 fracture

-0.78

cipline

-0.76

ilon

-0.73

 Norn

-0.73

ðĿ

-0.73

anship

-0.72

 theorem

-0.71

POSITIVE LOGITS

ads

2.13

 commercials

1.93

Ads

1.90

 advertising

1.86

 advertisements

1.85

 billboards

1.83

 Advertising

1.78

 advertis

1.76

 advertisers

1.73

 advertisement

1.68

Activations Density 0.399%

phrases indicating eligibility or ineligibility for advertising

references to eligibility for advertising

No Comments

No Known Activations