INDEX

Explanations

references to doors, such as holding the door open or closing the door

oai_token-act-pair · gpt-3.5-turbo

the word "door" and its variations in different contexts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ollah

-0.78

irtual

-0.73

lihood

-0.73

ivia

-0.71

ting

-0.70

ual

-0.69

TING

-0.68

ousand

-0.65

azeera

-0.65

milo

-0.65

POSITIVE LOGITS

bell

1.31

door

1.19

steps

1.14

 doors

1.06

doors

1.03

ways

0.99

 opener

0.97

 Door

0.93

 door

0.92

holes

0.90

Activations Density 0.035%

references to doors, such as holding the door open or closing the door

the word "door" and its variations in different contexts

No Comments

No Known Activations