INDEX

Explanations

mentions of the name "Affleck"

oai_token-act-pair · gpt-3.5-turbo

references to the name "Affleck" or related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

OPLE

-0.88

DragonMagazine

-0.84

GY

-0.77

âĹ¼

-0.73

netflix

-0.70

AAF

-0.69

ATOR

-0.68

Nusra

-0.67

 Dare

-0.66

NEWS

-0.64

POSITIVE LOGITS

idav

1.19

Aff

1.09

irmed

1.07

irmation

1.06

irm

1.04

leck

1.02

licted

1.02

irms

0.98

iliate

0.96

onso

0.95

Activations Density 0.005%

mentions of the name "Affleck"

references to the name "Affleck" or related terms

No Comments

No Known Activations