INDEX

Explanations

phrases related to computer programming and code

oai_token-act-pair · gpt-3.5-turbo

references to popular culture and specific entities in media, particularly movies and characters

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

ADVERTISEMENT

-0.52

android

-0.48

chool

-0.47

Rated

-0.46

tesy

-0.45

Jennifer

-0.45

Michelle

-0.45

ï¸ı

-0.44

ãĤ£

-0.44

ãĥ¼ãĥĨãĤ£

-0.44

POSITIVE LOGITS

 mosqu

0.56

 exting

0.54

eers

0.50

tyr

0.50

etheless

0.50

 Palestin

0.50

 bombed

0.48

 eleph

0.47

 cumbers

0.47

 Palest

0.47

Activations Density 1.883%

phrases related to computer programming and code

references to popular culture and specific entities in media, particularly movies and characters

No Comments

No Known Activations