INDEX

Explanations

references to the Android API version and application components

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

android support libraries

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

 propuesto

-0.53

ilarang

-0.49

 expuesto

-0.48

endregion

-0.47

 útil

-0.47

 nakalista

-0.46

 Infór

-0.46

 práctico

-0.46

 jugado

-0.45

 escuchado

-0.44

POSITIVE LOGITS

appcompat

0.66

bootstrapcdn

0.60

SupportActionBar

0.54

complexContent

0.52

 defaultstate

0.50

AnimationsModule

0.50

 uLocal

0.48

AppCompat

0.46

jface

0.46

 незавершена

0.45

Activations Density 0.001%

references to the Android API version and application components

android support libraries

No Comments

No Known Activations