INDEX

Explanations

phrases that encourage engagement and calls to action

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

Followed by the word "look"

np_acts-logits-general · gemini-2.0-flash

take a look

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

onite

-0.62

Искәрмәләр

-0.60

منابع

-0.58

bhan

-0.56

Дереккөздер

-0.54

 Emits

-0.54

CPtr

-0.53

moiselle

-0.53

executeUpdate

-0.53

verwijspagina

-0.53

POSITIVE LOGITS

 look

1.47

look

1.24

 Look

1.01

 LOOK

0.98

Look

0.96

 closer

0.84

 take

0.84

 looks

0.83

 peak

0.81

 have

0.80

Activations Density 0.151%

phrases that encourage engagement and calls to action

Followed by the word "look"

take a look

No Comments

No Known Activations