INDEX

Explanations

references to the concept of censorship

oai_token-act-pair · gpt-3.5-turbo

terms related to censorship

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

amac

-0.74

verty

-0.74

dden

-0.70

ndra

-0.70

 Lakes

-0.68

Dew

-0.68

arious

-0.68

ptoms

-0.67

gain

-0.67

ilater

-0.66

POSITIVE LOGITS

 censorship

1.06

 cens

1.04

 censor

0.95

 censored

0.95

 suppressing

0.72

 commissions

0.72

ourgeois

0.68

 selectively

0.67

creen

0.66

 blackout

0.66

Activations Density 0.022%

references to the concept of censorship

terms related to censorship

No Comments

No Known Activations