INDEX

Explanations

references to secrecy and hidden truths

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

 Paglinawan

-0.57

 extinguisher

-0.52

IRUS

-0.52

queryInterface

-0.50

TryGetValue

-0.49

 considérons

-0.48

ArrowToggle

-0.47

afone

-0.46

upassen

-0.46

tonode

-0.46

POSITIVE LOGITS

 hidden

1.67

 secretly

1.60

 behind

1.52

behind

1.44

hidden

1.38

Behind

1.38

Hidden

1.31

 Behind

1.30

 BEHIND

1.28

 Hidden

1.28

Activations Density 0.479%

references to secrecy and hidden truths

No Comments

No Known Activations