INDEX

Explanations

references to marriage-related terms, such as "groom" and "bride"

oai_token-act-pair · gpt-3.5-turbo

references to the roles of groom and bride in wedding contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

izoph

-0.74

itamin

-0.71

 Flavoring

-0.71

oids

-0.70

obin

-0.63

rael

-0.62

isd

-0.61

oid

-0.60

odium

-0.60

odic

-0.60

POSITIVE LOGITS

 groom

1.15

xual

1.12

stakes

0.91

eering

0.91

bones

0.86

 Bride

0.86

wra

0.80

asus

0.79

zilla

0.77

llers

0.77

Activations Density 0.010%

references to marriage-related terms, such as "groom" and "bride"

references to the roles of groom and bride in wedding contexts

No Comments

No Known Activations