INDEX

Explanations

mentions of addiction-related terms

oai_token-act-pair · gpt-3.5-turbo

terms related to addiction

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

Kov

-0.82

 Opera

-0.74

erial

-0.74

vard

-0.71

eor

-0.69

stand

-0.69

 Dayton

-0.67

 Compos

-0.65

FIR

-0.65

tell

-0.65

POSITIVE LOGITS

 addiction

1.05

 relapse

0.98

 addicted

0.94

 addict

0.93

 craving

0.91

 rehab

0.87

encies

0.85

 Addiction

0.84

olescent

0.83

icts

0.83

Activations Density 0.018%

mentions of addiction-related terms

terms related to addiction

No Comments

No Known Activations