INDEX

Explanations

references to e-books

oai_token-act-pair · gpt-3.5-turbo

mentions of e-books

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

 Pearce

-0.74

 wake

-0.66

 twitch

-0.64

Yin

-0.64

 Shank

-0.62

 Independence

-0.62

 theirs

-0.62

 Territories

-0.62

 dism

-0.61

 Hert

-0.61

POSITIVE LOGITS

books

1.37

poons

1.13

DragonMagazine

1.10

book

1.09

marks

1.09

 Journals

1.07

stores

1.05

acters

1.02

oldown

0.91

reads

0.86

Activations Density 0.007%

references to e-books

mentions of e-books

No Comments

No Known Activations