INDEX

Explanations

phrases relating to physical locations or positions

oai_token-act-pair · gpt-3.5-turbo

references to written or printed material

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

hov

-0.77

cosystem

-0.75

lig

-0.75

rients

-0.73

aclysm

-0.71

aughs

-0.71

ursion

-0.69

hedon

-0.69

utenberg

-0.69

arcity

-0.67

POSITIVE LOGITS

 screen

1.22

 billboards

1.20

screen

1.17

 forehead

1.16

 homepage

1.16

 walls

1.15

 underside

1.14

 sleeve

1.13

 page

1.13

 display

1.11

Activations Density 0.188%

phrases relating to physical locations or positions

references to written or printed material

No Comments

No Known Activations