INDEX

Explanations

titles and clickable text associated with user interface elements

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

tooltip definitions

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

 ویکی‌پدیا

-0.68

########.

-0.67

InputBorder

-0.65

Tembelea

-0.63

IntoConstraints

-0.60

nloa

-0.57

FormTagHelper

-0.56

verwijspagina

-0.55

balleur

-0.55

Cyfeiriadau

-0.54

POSITIVE LOGITS

setToolTipText

0.40

setToolTip

0.38

bitField

0.36

Tooltip

0.35

Архівовано

0.34

 thiệu

0.34

collant

0.33

tooltip

0.31

անում

0.31

FULL

0.31

Activations Density 0.008%

titles and clickable text associated with user interface elements

tooltip definitions

No Comments

No Known Activations