INDEX

Explanations

phrases related to doors

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "door"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Hots

-0.71

enegger

-0.69

TY

-0.67

udeb

-0.67

ovych

-0.65

azeera

-0.65

lihood

-0.64

ontent

-0.64

TING

-0.64

ciating

-0.62

POSITIVE LOGITS

bell

1.27

door

1.23

steps

1.09

holes

1.07

 doors

1.02

frame

0.96

doors

0.94

 door

0.94

 Door

0.91

hole

0.90

Activations Density 0.015%

phrases related to doors

occurrences of the word "door"

No Comments

No Known Activations