INDEX

Explanations

mentions of the word "book."

oai_token-act-pair · gpt-3.5-turbo

mentions of books

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

ilitary

-0.75

 Bots

-0.72

sembly

-0.68

xon

-0.67

 distant

-0.65

 twitch

-0.64

cffff

-0.64

Yin

-0.64

 Lumpur

-0.64

 VIDEOS

-0.62

POSITIVE LOGITS

stores

1.49

seller

1.27

marks

1.27

shop

1.14

marked

1.11

cases

1.09

worms

1.08

worm

1.05

book

1.03

keeping

1.02

Activations Density 0.038%

mentions of the word "book."

mentions of books

No Comments

No Known Activations