INDEX

Explanations

instances where someone is being advised or reminded to pay attention to something

oai_token-act-pair · gpt-3.5-turbo

phrases emphasizing paying attention

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

rie

-0.73

yre

-0.64

iHUD

-0.63

lua

-0.62

\\\\\\\\

-0.61

venge

-0.58

\\\\

-0.58

atu

-0.57

 Royale

-0.57

apons

-0.57

POSITIVE LOGITS

 attent

0.90

to

0.74

 NOTICE

0.73

olescent

0.68

é¾įå¥ĳå£«

0.66

idious

0.66

 closely

0.65

itness

0.65

 attention

0.64

ibilities

0.64

Activations Density 0.050%

instances where someone is being advised or reminded to pay attention to something

phrases emphasizing paying attention

No Comments

No Known Activations