INDEX

Explanations

proper nouns related to brands and media outlets

oai_token-act-pair · gpt-3.5-turbo

references to specific media entities and characters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ener

-0.79

lyak

-0.78

aeda

-0.74

stress

-0.72

gments

-0.71

cius

-0.71

arily

-0.70

inen

-0.69

eric

-0.68

ars

-0.68

POSITIVE LOGITS

 Atlantis

0.83

AMA

0.81

vier

0.80

 Alley

0.75

terness

0.71

 Boards

0.71

Wem

0.71

 Royale

0.71

Peb

0.70

Panel

0.68

Activations Density 0.100%

proper nouns related to brands and media outlets

references to specific media entities and characters

No Comments

No Known Activations