INDEX

Explanations

words related to substance abuse and illegal drug use

oai_token-act-pair · gpt-3.5-turbo

references to substance use and abuse

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

Temp

-0.87

Solar

-0.86

Hug

-0.81

Travel

-0.78

AIR

-0.77

RECT

-0.76

Wind

-0.76

Bird

-0.75

AUT

-0.75

FOX

-0.75

POSITIVE LOGITS

 methamphetamine

1.07

 prohibition

1.04

 addiction

1.00

 disorders

1.00

 overdose

0.99

 drugs

0.99

 cartel

0.98

 Prohibition

0.97

 cocaine

0.97

 overdoses

0.97

Activations Density 0.086%

words related to substance abuse and illegal drug use

references to substance use and abuse

No Comments

No Known Activations