INDEX

Explanations

references to criminal activities or offenses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ReusableCell

-0.60

 Superhost

-0.60

IHC

-0.59

 MainAxisSize

-0.59

NewUrlParser

-0.54

UnitTesting

-0.53

 Correspondence

-0.52

HomeAsUpEnabled

-0.52

complexContent

-0.51

 Seeder

-0.51

POSITIVE LOGITS

 theft

1.98

 stolen

1.97

 thief

1.91

 thieves

1.85

 stealing

1.78

 robbery

1.78

 steal

1.77

 thefts

1.74

 Theft

1.68

 stole

1.65

Activations Density 0.445%

references to criminal activities or offenses

No Comments

No Known Activations