INDEX

Explanations

programming-related keywords, particularly import and include statements

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

code includes and imports

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

 Justiça

-0.32

 meines

-0.29

jeta

-0.29

 desee

-0.29

beeld

-0.28

 being

-0.28

 Preferencias

-0.28

UNDA

-0.27

addValue

-0.27

 jurídica

-0.27

POSITIVE LOGITS

帖最后由

0.72

 المعيارى

0.72

ChildScrollView

0.63

RTLR

0.61

Hentet

0.60

__(/*!

0.56

 ComVisible

0.55

 Administrativna

0.54

 متعلقه

0.53

BASEPATH

0.52

Activations Density 0.006%

programming-related keywords, particularly import and include statements

code includes and imports

No Comments

No Known Activations