INDEX

Explanations

mentions of the word "torture" in documents

oai_token-act-pair · gpt-3.5-turbo

mentions of torture

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ership

-0.76

kj

-0.66

 Financial

-0.65

aunder

-0.64

ovember

-0.64

irc

-0.63

Merit

-0.63

amins

-0.62

ijk

-0.62

iod

-0.62

POSITIVE LOGITS

 torture

1.08

 captives

0.81

 torment

0.80

 detainees

0.80

isen

0.79

 tortured

0.78

 interrog

0.75

APE

0.74

 Tort

0.72

apons

0.71

Activations Density 0.008%

mentions of the word "torture" in documents

mentions of torture

No Comments

No Known Activations