INDEX

Explanations

mentions of readers

oai_token-act-pair · gpt-3.5-turbo

mentions of "readers" 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 @ 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

 Parenthood

-0.75

asia

-0.70

 Pharma

-0.67

apes

-0.64

 manslaughter

-0.62

 Species

-0.62

 wreck

-0.61

xon

-0.61

redo

-0.60

ayne

-0.60

POSITIVE LOGITS

 readers

1.27

 Readers

1.16

 reader

1.16

hip

1.13

DragonMagazine

0.93

 Reader

0.91

reader

0.91

lisher

0.85

worms

0.83

geist

0.82

Activations Density 0.009%

mentions of readers

mentions of "readers" in various contexts

No Comments

No Known Activations