INDEX

Explanations

proper nouns related to a specific figure or name, namely "Bin Laden."

oai_token-act-pair · gpt-3.5-turbo

mentions of "Bin Laden."

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

anwhile

-0.81

sburgh

-0.79

ITH

-0.74

mble

-0.72

Premium

-0.68

pter

-0.67

 Raise

-0.66

 Decay

-0.65

ESA

-0.64

ULT

-0.63

POSITIVE LOGITS

ocular

1.41

Bin

1.26

 Laden

1.24

omial

1.24

nington

1.03

bin

1.00

thood

0.87

aries

0.86

jamin

0.83

oha

0.81

Activations Density 0.003%

proper nouns related to a specific figure or name, namely "Bin Laden."

mentions of "Bin Laden."

No Comments

No Known Activations