INDEX

Explanations

terms related to fraudulent activities

oai_token-act-pair · gpt-3.5-turbo

words and phrases related to fraudulent activities and deception

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

baugh

-0.65

ŃĶ

-0.64

âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ

-0.62

 Ange

-0.61

 Difference

-0.60

 Quartz

-0.58

ISO

-0.58

 Despair

-0.57

 Afric

-0.56

 Colon

-0.56

POSITIVE LOGITS

ulent

1.66

ulence

1.63

sters

1.26

ster

1.18

ul

1.04

 perpetrated

0.99

ulators

0.97

ulus

0.96

ulously

0.95

inals

0.93

Activations Density 0.034%

terms related to fraudulent activities

words and phrases related to fraudulent activities and deception

No Comments

No Known Activations