INDEX

Explanations

words related to crime

oai_token-act-pair · gpt-3.5-turbo

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

 Dangerous

-0.65

 Lisbon

-0.64

CAR

-0.64

Constructed

-0.60

grades

-0.58

 Senegal

-0.58

 Schumer

-0.58

 Mandela

-0.58

eers

-0.57

lihood

-0.57

POSITIVE LOGITS

onial

1.03

inals

1.02

ming

1.00

inatory

0.96

rim

0.94

inally

0.93

inating

0.89

etric

0.89

med

0.88

shaw

0.88

Activations Density 0.037%

words related to crime

No Comments

No Known Activations