INDEX

Explanations

instances of fraudulent and deceptive behavior

oai_token-act-pair · gpt-3.5-turbo

terms related to fraudulent and deceptive practices

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

arium

-0.77

nat

-0.75

hung

-0.74

area

-0.73

mun

-0.73

bur

-0.72

alist

-0.72

hed

-0.72

resent

-0.71

raq

-0.70

POSITIVE LOGITS

 fraudulent

0.88

 scam

0.85

 unsuspecting

0.85

 fraud

0.84

 scams

0.82

 dece

0.82

 deceive

0.78

 manipulative

0.77

 deception

0.76

 cheat

0.75

Activations Density 0.025%

instances of fraudulent and deceptive behavior

terms related to fraudulent and deceptive practices

No Comments

No Known Activations