INDEX

Explanations

dates and titles

oai_token-act-pair · gpt-3.5-turbo

elements related to trivia or informational content

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

merce

-0.83

amacare

-0.79

hello

-0.74

bang

-0.73

answer

-0.72

brainer

-0.72

olding

-0.71

thouse

-0.69

EMA

-0.69

arching

-0.68

POSITIVE LOGITS

 Edit

1.03

Trivia

1.01

Contents

0.93

Appearances

0.93

".[

0.92

 flashback

0.91

 canon

0.88

Gallery

0.86

,[

0.84

edit

0.83

Activations Density 0.735%

dates and titles

elements related to trivia or informational content

No Comments

No Known Activations