INDEX

Explanations

references to censorship and related controversies

oai_token-act-pair · gpt-3.5-turbo

mentions of censorship and its related concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Lakes

-0.81

amac

-0.79

Leary

-0.78

ths

-0.78

ptoms

-0.73

thritis

-0.72

nie

-0.72

thus

-0.72

ndra

-0.72

acts

-0.71

POSITIVE LOGITS

 censorship

1.45

 censor

1.33

 cens

1.21

 censored

1.13

 suppression

0.82

 suppressing

0.78

ourgeois

0.77

 cutter

0.77

 disadvant

0.75

 blackout

0.75

Activations Density 0.012%

references to censorship and related controversies

mentions of censorship and its related concepts

No Comments

No Known Activations