INDEX

Explanations

information related to law enforcement and security measures

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

terms related to terrorism and the measures taken against it

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_post

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

paio

-0.93

istrate

-0.81

 Garfield

-0.79

ellen

-0.77

 Starr

-0.77

ripp

-0.75

 Veter

-0.74

Â¢

-0.74

fred

-0.73

Cola

-0.72

POSITIVE LOGITS

 extremist

1.42

 radical

1.40

 jihadist

1.40

 extremism

1.36

 jihad

1.30

 Islamist

1.28

 ISIL

1.26

 extremists

1.24

 mosques

1.24

 recruitment

1.22

Activations Density 0.461%

information related to law enforcement and security measures

terms related to terrorism and the measures taken against it

No Comments

No Known Activations