INDEX

Explanations

phrases related to online privacy notices

oai_token-act-pair · gpt-3.5-turbo

references to privacy policies and notices

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 scapego

-0.56

 retri

-0.55

 corrid

-0.54

 patriarch

-0.52

 blat

-0.50

 elev

-0.50

 coord

-0.50

 chuck

-0.49

 resorted

-0.49

 sucker

-0.48

POSITIVE LOGITS

 Could

0.75

ilty

0.64

iphate

0.60

Domain

0.53

ylon

0.52

 Errors

0.52

iverse

0.52

Privacy

0.51

antha

0.49

abetes

0.49

Activations Density 0.014%

phrases related to online privacy notices

references to privacy policies and notices

No Comments

No Known Activations