INDEX

Explanations

structural elements and keywords used in programming, particularly in user interface contexts

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

web/app display properties

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

ren

-0.31

jface

-0.28

ngrx

-0.27

 InputDecoration

-0.26

 nominate

-0.26

 obě

-0.25

 eaten

-0.24

Persist

-0.24

 JADX

-0.24

 ozna

-0.23

POSITIVE LOGITS

yntaxException

0.78

ſelf

0.64

 BoxFit

0.63

 pleaſure

0.62

 zwiſchen

0.60

 houſe

0.60

 poffible

0.59

 myſelf

0.59

ſelves

0.59

__))

0.58

Activations Density 0.008%

structural elements and keywords used in programming, particularly in user interface contexts

web/app display properties

No Comments

No Known Activations