INDEX

Explanations

information related to investigations or claims

oai_token-act-pair · gpt-3.5-turbo

instances of legal or investigative scenarios involving claims and accusations

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

iku

-0.55

iona

-0.55

 bask

-0.54

ggles

-0.53

gri

-0.53

uitive

-0.53

 incent

-0.52

 partName

-0.52

auntlet

-0.51

vironments

-0.50

POSITIVE LOGITS

 inappropriately

0.71

 prostitutes

0.70

 Kislyak

0.68

nikov

0.66

 illegally

0.66

 improperly

0.66

 prank

0.65

 paed

0.64

 Paddock

0.63

 Bulgar

0.63

Activations Density 1.592%

information related to investigations or claims

instances of legal or investigative scenarios involving claims and accusations

No Comments

No Known Activations