INDEX

Explanations

terms related to criminal activities, specifically gang-related

oai_token-act-pair · gpt-3.5-turbo

mentions of gangs and gang-related 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

Not in Any Lists

No Comments

Negative Logits

etsk

-0.78

hower

-0.77

andum

-0.68

 Farn

-0.67

eem

-0.67

*/(

-0.65

 concess

-0.64

 Dayton

-0.63

OPLE

-0.63

Ecc

-0.62

POSITIVE LOGITS

sta

1.47

sters

1.31

ster

1.28

busters

1.24

bang

1.23

rene

1.08

members

0.96

buster

0.94

member

0.90

regate

0.89

Activations Density 0.036%

terms related to criminal activities, specifically gang-related

mentions of gangs and gang-related activities

No Comments

No Known Activations