INDEX

Explanations

mentions of law enforcement agencies and related activities

oai_token-act-pair · gpt-3.5-turbo

mentions of police-related activities and events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

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

 Remastered

-0.82

anyl

-0.75

 Lear

-0.73

Minnesota

-0.72

 Decay

-0.70

 Whitney

-0.69

 Lilly

-0.69

Virginia

-0.68

 Idaho

-0.68

maxwell

-0.67

POSITIVE LOGITS

 officers

1.10

 colonel

1.06

 raided

1.05

 officer

1.02

 crackdown

1.02

 forces

1.01

 chief

0.99

 reinforcements

0.99

 armoured

0.94

 detained

0.94

Activations Density 0.123%

mentions of law enforcement agencies and related activities

mentions of police-related activities and events

No Comments

No Known Activations