INDEX

Explanations

references to web technologies or programming-related terms

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

static content and lfs

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

 Administrativna

-0.61

tagHelperRunner

-0.55

Espèce

-0.55

endsection

-0.55

reportWebVitals

-0.54

yii

-0.54

exitRule

-0.53

 hObject

-0.53

 snippetHide

-0.53

 BorderRadius

-0.52

POSITIVE LOGITS

static

0.94

 static

0.68

Static

0.63

STATIC

0.60

 Static

0.56

 STATIC

0.54

静态

0.49

twimg

0.48

imageshack

0.47

стати

0.41

Activations Density 0.000%

references to web technologies or programming-related terms

static content and lfs

No Comments

No Known Activations