INDEX

Explanations

law enforcement-related terms

oai_token-act-pair · gpt-3.5-turbo

terms related to law enforcement and police activity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

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

ebted

-0.84

FORE

-0.81

Sov

-0.77

hend

-0.77

rencies

-0.73

åĮ

-0.72

apore

-0.72

ppo

-0.70

Reloaded

-0.69

jri

-0.69

POSITIVE LOGITS

 officers

1.15

 patrolling

1.09

 stationed

1.05

 Officers

1.03

 officer

1.03

 patrol

1.01

 dispatcher

1.00

 SWAT

0.94

 sergeant

0.93

 dispatched

0.91

Activations Density 0.125%

law enforcement-related terms

terms related to law enforcement and police activity

No Comments

No Known Activations