INDEX

Explanations

terms related to user input and interaction within software applications

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

automatic or automatically

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

TargetException

-0.43

RenderAtEndOf

-0.41

EDEFAULT

-0.40

 aikaa

-0.39

gdx

-0.38

 linkovi

-0.38

ValueStyle

-0.36

sikt

-0.34

 Yugos

-0.32

zionalità

-0.32

POSITIVE LOGITS

 automatically

0.57

 automaticamente

0.55

 automatisch

0.49

automatically

0.47

TextWatcher

0.46

 automatica

0.46

 CommonModule

0.45

自動

0.45

 Automatically

0.44

 자동

0.44

Activations Density 0.158%

terms related to user input and interaction within software applications

automatic or automatically

No Comments

No Known Activations