INDEX

Explanations

programming language constructs and references, particularly related to class and function definitions

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

Code, programming, or software-related terms

np_acts-logits-general · gemini-2.0-flash

multi-language code tokens

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

TextBoxColumn

-0.57

 Handmaid

-0.56

OkHttpClient

-0.53

WithIdentifier

-0.53

fusc

-0.49

InjectMocks

-0.47

σθαι

-0.47

 much

-0.45

routeProvider

-0.45

 RemoteException

-0.44

POSITIVE LOGITS

تقاوى

0.81

 الحره

0.78

存于互联网档案馆

0.75

hyrchwyd

0.69

AndEndTag

0.68

 Chwiliwch

0.66

 utafitiHapana

0.66

دانشنامهٔ

0.63

########.

0.63

 مرئيه

0.61

Activations Density 0.220%

programming language constructs and references, particularly related to class and function definitions

Code, programming, or software-related terms

multi-language code tokens

No Comments

No Known Activations