INDEX

Explanations

the word "sense" when used in a context of logic or reason

oai_token-act-pair · gpt-3.5-turbo

phrases that denote understanding or logical reasoning

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

enne

-0.69

astered

-0.68

venge

-0.67

ewitness

-0.67

berus

-0.63

ching

-0.63

vol

-0.63

inent

-0.63

inous

-0.63

aunting

-0.63

POSITIVE LOGITS

 partName

0.78

why

0.76

INESS

0.76

soDeliveryDate

0.76

é¾įå¥ĳå£«

0.74

 Eater

0.74

WHY

0.69

imaru

0.69

DragonMagazine

0.68

Compat

0.68

Activations Density 0.028%

the word "sense" when used in a context of logic or reason

phrases that denote understanding or logical reasoning

No Comments

No Known Activations