INDEX

Explanations

names of people or places

oai_token-act-pair · gpt-3.5-turbo

the names of individuals, particularly with the surname 'Donnelly' and references to mourning

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ħĭ

-0.78

scenes

-0.77

yrinth

-0.74

erion

-0.70

ĺħ

-0.70

RAM

-0.68

SY

-0.68

erate

-0.68

devices

-0.68

lasses

-0.68

POSITIVE LOGITS

 Donn

0.99

 Beir

0.88

Ò

0.87

ners

0.86

acha

0.78

spir

0.78

Í

0.76

ning

0.76

izontal

0.75

izont

0.73

Activations Density 0.024%

names of people or places

the names of individuals, particularly with the surname 'Donnelly' and references to mourning

No Comments

No Known Activations