INDEX

Explanations

locations

oai_token-act-pair · gpt-3.5-turbo

references to geographic locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

DragonMagazine

-0.77

agency

-0.67

rive

-0.64

MJ

-0.62

itus

-0.61

sein

-0.61

©¶æ

-0.61

vr

-0.61

ĪĴ

-0.60

itary

-0.60

POSITIVE LOGITS

where

1.33

 where

1.29

 Africa

1.22

 Europe

1.21

 Austria

1.20

 Belgium

1.18

 Italy

1.18

 Spain

1.17

 England

1.17

 Switzerland

1.17

Activations Density 0.228%

locations

references to geographic locations

No Comments

No Known Activations