INDEX

Explanations

phrases related to user consent and communication preferences

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

marketing and unsolicited communication

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

 Valdés

-0.42

<?

-0.39

 defaultstate

-0.39

 relever

-0.37

 Núñez

-0.37

 escultura

-0.36

ukone

-0.35

 contemporaine

-0.34

انيف

-0.34

 fark

-0.34

POSITIVE LOGITS

ActionCreators

0.52

PerformLayout

0.51

:✨

0.51

 setEmail

0.49

EFIAPI

0.45

 kasarigan

0.45

ItemBackground

0.44

 vervol

0.43



0.42

 <<<<<<<<<<<<<<

0.42

Activations Density 0.012%

phrases related to user consent and communication preferences

marketing and unsolicited communication

No Comments

No Known Activations