INDEX

Explanations

phrases related to spam and unsolicited messages

oai_token-act-pair · gpt-3.5-turbo

references to spam and scams

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hani

-1.17

 Borders

-0.70

IST

-0.67

 Cheong

-0.67

 Vernon

-0.65

Cel

-0.65

ĸļ

-0.63

 supervised

-0.63

 Dayton

-0.61

 Templ

-0.59

POSITIVE LOGITS

ming

1.40

ulent

0.97

ulus

0.93

my

0.93

mer

0.91

inator

0.89

mers

0.88

 spam

0.87

trap

0.83

ulence

0.82

Activations Density 0.023%

phrases related to spam and unsolicited messages

references to spam and scams

No Comments

No Known Activations