INDEX

Explanations

requests or polite pleas for action

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

"Please" followed by a verb

np_acts-logits-general · gemini-2.0-flash

polite requests

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

WithIOException

-0.89

XmlAccessType

-0.81

Personensuche

-0.74

IsContent

-0.73

χεία

-0.71

 estekak

-0.67

 հղումներ

-0.67

EndContext

-0.65

كويكب

-0.63

POSITIVE LOGITS

don

0.88

 please

0.81

Please

0.78

 Please

0.76

 note

0.75

let

0.73

please

0.71

 PLEASE

0.69

 Kindly

0.68

do

0.66

Activations Density 0.041%

requests or polite pleas for action

"Please" followed by a verb

polite requests

No Comments

No Known Activations