INDEX

Explanations

instances of fraud or deceitful activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 itſelf

-0.73

Efq

-0.70

 fortific

-0.67

 intStringLen

-0.67

ValueStyle

-0.66

TagMode

-0.66

 myſelf

-0.64

 erec

-0.63

 ſind

-0.63

 <=",

-0.63

POSITIVE LOGITS

 scam

1.06

 scammed

0.96

 fooled

0.95

 scams

0.94

 defraud

0.88

 fraud

0.88

 deceived

0.87

 deceive

0.83

 fraude

0.83

 cheated

0.82

Activations Density 0.293%

instances of fraud or deceitful activities

No Comments

No Known Activations