INDEX

Explanations

dynamic interactions or changes in automated systems

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

before technical terms or code snippets

np_acts-logits-general · gemini-2.0-flash

user intent or action

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

KURZBESCHREIBUNG

-0.42

 fuper

-0.41

openhauer

-0.40

RenderAtEndOf

-0.40

 raiſ

-0.40

 Inſ

-0.40

 poffible

-0.40

 occaf

-0.38

 Kontrol

-0.38

 beſ

-0.36

POSITIVE LOGITS

 CommonModule

0.48

setOnItem

0.46

setupUi

0.44

 BrowserModule

0.44

ChildScrollView

0.43

TextWatcher

0.43

FailureListener

0.43

openConnection

0.42

ValueGenerated

0.42

FieldError

0.41

Activations Density 0.639%

dynamic interactions or changes in automated systems

before technical terms or code snippets

user intent or action

No Comments

No Known Activations