INDEX

Explanations

mentions of financial wrongdoing or criminal deception

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "fraud" and its related context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ŃĶ

-0.73

âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ

-0.73

UTC

-0.68

 Idle

-0.63

 Quartz

-0.62

 Candle

-0.62

STD

-0.61

%"

-0.61

Benz

-0.61

 Fitness

-0.59

POSITIVE LOGITS

ulence

1.47

ulent

1.46

sters

1.06

ster

1.01

ously

0.93

ulously

0.89

 fraud

0.89

 perpetrated

0.87

inals

0.87

ulus

0.85

Activations Density 0.019%

mentions of financial wrongdoing or criminal deception

occurrences of the word "fraud" and its related context

No Comments

No Known Activations