INDEX

Explanations

phrases related to people's names, especially the name "Heidi"

oai_token-act-pair · gpt-3.5-turbo

names and terms related to individuals and their actions or states

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

atri

-0.75

ovych

-0.73

 externalToEVAOnly

-0.72

xia

-0.71

ambers

-0.70

atile

-0.70

 arrivals

-0.69

anyl

-0.69

hene

-0.69

atics

-0.68

POSITIVE LOGITS

defense

0.83

attack

0.80

aminer

0.76

offensive

0.70

 Uriel

0.70

 wart

0.69

blow

0.69

 Heidi

0.68

wered

0.67

Miy

0.65

Activations Density 0.019%

phrases related to people's names, especially the name "Heidi"

names and terms related to individuals and their actions or states

No Comments

No Known Activations