INDEX

Explanations

prepositions followed by words indicating relationships or emotions

oai_token-act-pair · gpt-3.5-turbo

expressions related to love, relationships, and moral actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

interstitial

-0.93

endiary

-0.69

 Donation

-0.68

uran

-0.66

ulations

-0.63

arbon

-0.63

 millenn

-0.62

buster

-0.62

vertisement

-0.60

 prepar

-0.60

POSITIVE LOGITS

him

1.51

me

1.48

us

1.46

Him

1.33

 others

1.33

 themselves

1.30

him

1.29

 oneself

1.28

 yourself

1.25

 herself

1.24

Activations Density 0.356%

prepositions followed by words indicating relationships or emotions

expressions related to love, relationships, and moral actions

No Comments

No Known Activations