INDEX

Explanations

words, phrases, or proper nouns related to a specific entity or brand that starts with the letters "we"

oai_token-act-pair · gpt-3.5-turbo

the word "we" in various contexts throughout the document

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

Posts

-0.66

 firsthand

-0.64

 gratification

-0.59

 srfAttach

-0.59

 Seat

-0.58

 Crush

-0.56

 Kush

-0.55

 sucker

-0.54

 Highlights

-0.53

Lep

-0.53

POSITIVE LOGITS

bsite

1.42

athered

1.38

lder

1.32

asel

1.28

eding

1.27

avers

1.24

eping

1.23

arers

1.21

ldon

1.20

aving

1.20

Activations Density 0.088%

words, phrases, or proper nouns related to a specific entity or brand that starts with the letters "we"

the word "we" in various contexts throughout the document

No Comments

No Known Activations