INDEX

Explanations

the word "observers" with high activation values

oai_token-act-pair · gpt-3.5-turbo

mentions of observers or observational roles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

street

-0.67

eways

-0.67

Hom

-0.67

False

-0.66

Enough

-0.63

rax

-0.62

Bio

-0.62

sis

-0.62

Customer

-0.62

eating

-0.61

POSITIVE LOGITS

 observers

1.28

 observer

1.22

wat

1.03

acers

0.92

 observing

0.87

auts

0.82

opol

0.81

 onlook

0.79

 "$:/

0.78

 rejoice

0.78

Activations Density 0.013%

the word "observers" with high activation values

mentions of observers or observational roles

No Comments

No Known Activations