INDEX

Explanations

queries and requests for information or assistance

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

Followed by a preposition or comma

np_acts-logits-general · gemini-2.0-flash

please and requests

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 intptr

-0.81

存于互联网档案馆

-0.79

 Мексичка

-0.72

 Administrativna

-0.64

aarrggbb

-0.62

extAlignment

-0.58

彿

-0.56

 unknownFields

-0.56

//*/

-0.56

InjectAttribute

-0.55

POSITIVE LOGITS

 please

1.97

please

1.68

 Please

1.61

Please

1.55

 PLEASE

1.53

pls

1.39

 hãy

1.33

请

1.32

plz

1.30

PLEASE

1.27

Activations Density 0.373%

queries and requests for information or assistance

Followed by a preposition or comma

please and requests

No Comments

No Known Activations