INDEX

Explanations

words related to books or publications

oai_token-act-pair · gpt-3.5-turbo

references to various types of books or publication formats

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 kindred

-0.64

 Pearce

-0.64

 hell

-0.62

 cease

-0.60

 Griffith

-0.57

 Enlightenment

-0.57

itcher

-0.57

 skepticism

-0.57

pol

-0.56

 microw

-0.56

POSITIVE LOGITS

book

1.42

marks

1.22

books

1.20

stores

1.13

cloth

1.01

marked

0.99

mark

0.97

shop

0.96

worm

0.95

Book

0.93

Activations Density 0.013%

words related to books or publications

references to various types of books or publication formats

No Comments

No Known Activations