INDEX

Explanations

mentions of espionage or spies

oai_token-act-pair · gpt-3.5-turbo

references to espionage and spies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

xual

-0.88

à¼

-0.80

ĸļ

-0.78

esville

-0.78

orical

-0.70

opathic

-0.70

usable

-0.70

clus

-0.69

Merit

-0.68

requisite

-0.68

POSITIVE LOGITS

 spies

0.90

spy

0.89

sonian

0.89

oleon

0.88

glass

0.82

 spoof

0.81

 satellites

0.81

 spying

0.76

OSH

0.75

 roundup

0.72

Activations Density 0.012%

mentions of espionage or spies

references to espionage and spies

No Comments

No Known Activations