INDEX

Explanations

terms related to fraudulent schemes or deception

oai_token-act-pair · gpt-3.5-turbo

references to scams and fraudulent activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

âĪ

-0.90

 Elements

-0.81

ef

-0.74

fif

-0.71

Ax

-0.69

idge

-0.69

ridge

-0.65

 defined

-0.64

table

-0.64

Equ

-0.64

POSITIVE LOGITS

 scam

3.84

 scams

2.87

 hoax

1.69

 fraud

1.68

 Fraud

1.53

 prank

1.48

 racket

1.43

 fraudulent

1.40

 spam

1.37

raud

1.32

Activations Density 0.022%

terms related to fraudulent schemes or deception

references to scams and fraudulent activities

No Comments

No Known Activations