INDEX

Explanations

information related to criminal investigations, incidents, and law enforcement activities

oai_token-act-pair · gpt-3.5-turbo

phrases related to criminal investigations and law enforcement activities

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

inav

-0.81

 persu

-0.77

 disadvant

-0.75

 fulfil

-0.72

 shrew

-0.72

 reckoning

-0.69

 ingen

-0.68

 defeats

-0.67

 monopoly

-0.67

 systematically

-0.66

POSITIVE LOGITS

Investigators

1.11

APD

1.07

 detectives

1.01

Police

1.01

 Investigators

1.00

Stay

0.98

Alert

0.95

Authorities

0.95

Detect

0.91

police

0.91

Activations Density 0.780%

information related to criminal investigations, incidents, and law enforcement activities

phrases related to criminal investigations and law enforcement activities

No Known Activations