INDEX

Explanations

information related to police involvement in various incidents

oai_token-act-pair · gpt-3.5-turbo

mentions of law enforcement or police-related activities

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

xual

-0.87

opian

-0.80

yip

-0.79

osi

-0.76

pread

-0.72

minecraft

-0.68

ocene

-0.68

tem

-0.68

ãĤ¦

-0.67

POSITIVE LOGITS

 officers

1.20

 departments

1.12

 officer

1.09

 Officers

0.97

 department

0.96

 commissioner

0.95

 chiefs

0.92

 superintendent

0.91

 chief

0.87

 headquarters

0.86

Activations Density 0.068%

information related to police involvement in various incidents

mentions of law enforcement or police-related activities

No Comments

No Known Activations