INDEX

Explanations

mentions of police-related terms and activities

oai_token-act-pair · gpt-3.5-turbo

mentions of law enforcement or police activity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

opian

-0.90

xual

-0.83

osi

-0.79

owship

-0.74

thal

-0.73

cast

-0.73

arial

-0.73

yip

-0.72

igible

-0.70

perfect

-0.70

POSITIVE LOGITS

 officers

1.16

 departments

1.11

 officer

0.97

 Officers

0.96

 blot

0.93

 Commissioner

0.93

 Superintendent

0.92

 investigators

0.92

 commissioner

0.91

 scanners

0.91

Activations Density 0.060%

mentions of police-related terms and activities

mentions of law enforcement or police activity

No Comments

No Known Activations