INDEX

Explanations

mentions of the word "Cat" and related terms like "California", "Cathedral", and "catalog"

oai_token-act-pair · gpt-3.5-turbo

references to "Cat" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

mble

-0.84

htt

-0.78

gur

-0.75

hower

-0.71

 Seym

-0.69

FontSize

-0.65

uden

-0.65

 Vaugh

-0.65

byter

-0.63

assetsadobe

-0.63

POSITIVE LOGITS

aclysm

1.50

apult

1.17

alogue

1.13

heter

1.12

alyst

1.09

cher

1.07

hedral

1.06

cat

1.05

alog

1.03

chers

0.97

Activations Density 0.011%

mentions of the word "Cat" and related terms like "California", "Cathedral", and "catalog"

references to "Cat" in various contexts

No Comments

No Known Activations