INDEX

Explanations

terms related to law enforcement activities and legal violations, especially regarding drunk driving

oai_token-act-pair · gpt-3.5-turbo

references to law enforcement actions and violations related to driving offenses

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

sonian

-0.91

natureconservancy

-0.83

amorph

-0.79

isSpecialOrderable

-0.77

phil

-0.77

Tea

-0.76

Reviewer

-0.75

guided

-0.74

shr

-0.74

itcher

-0.72

POSITIVE LOGITS

 arrests

1.03

 homicide

1.01

 officer

0.98

 homicides

0.97

 arrest

0.96

 burglary

0.95

 felon

0.94

 LAPD

0.94

 manslaughter

0.94

©¶æ¥µ

0.92

Activations Density 1.169%

terms related to law enforcement activities and legal violations, especially regarding drunk driving

references to law enforcement actions and violations related to driving offenses

No Comments

No Known Activations