INDEX

Explanations

phrases related to illegal actions and laws

oai_token-act-pair · gpt-3.5-turbo

references to illegal activities and violations of law

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

yip

-0.99

itiveness

-0.83

antage

-0.82

voy

-0.79

ngth

-0.77

vell

-0.71

stretched

-0.71

 resil

-0.69

erella

-0.66

angular

-0.66

POSITIVE LOGITS

SpaceEngineers

0.94

ADA

0.86

 punishable

0.79

 QuÃ©

0.75

taboola

0.75

VPN

0.73

 Proposition

0.73

ãĥīãĥ©

0.72

ãĤ®

0.71

acea

0.68

Activations Density 0.382%

phrases related to illegal actions and laws

references to illegal activities and violations of law

No Comments

No Known Activations