INDEX

Explanations

phrases related to Johns Hopkins University

oai_token-act-pair · gpt-3.5-turbo

mentions of "Johns" or references to Johns Hopkins University

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

mble

-0.82

BOOK

-0.74

Spread

-0.72

 deleting

-0.70

ãĤ§

-0.66

AUT

-0.65

Examples

-0.63

PLA

-0.62

leased

-0.62

holding

-0.62

POSITIVE LOGITS

 Johns

1.34

 Hopkins

1.31

sonian

1.03

kins

0.92

arton

0.83

istry

0.79

ston

0.79

keye

0.75

otom

0.74

algia

0.74

Activations Density 0.002%

phrases related to Johns Hopkins University

mentions of "Johns" or references to Johns Hopkins University

No Comments

No Known Activations