INDEX

Explanations

elements related to programming structures and resources

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

Foreign and technical words

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

complexContent

-0.67

 AssemblyTitle

-0.55

TintMode

-0.53

ChildScrollView

-0.52

 initComponents

-0.52

MigrationBuilder

-0.52

UnsafeEnabled

-0.48

HideFlags

-0.48

 useAppContext

-0.48

 Signalez

-0.47

POSITIVE LOGITS

⅘

0.46

 bebés

0.42

 frère

0.40

 chère

0.39

åter

0.39

 sınır

0.38

 incluido

0.38

 niñas

0.37

baby

0.37

DropColumn

0.36

Activations Density 0.081%

elements related to programming structures and resources

Foreign and technical words

No Comments

No Known Activations