INDEX

Explanations

references to graphical layout components in code or diagrams

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

panel and bottom

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

tableFuture

-0.56

StructEnd

-0.51

 FileReader

-0.51

 المعيارى

-0.50

Примечания

-0.49

 FormData

-0.49

Зноскі

-0.49

 للمعارف

-0.48

ագրություններ

-0.48

spun

-0.47

POSITIVE LOGITS

windowFixed

0.49

Parcelize

0.48

MigrationBuilder

0.45

httphttps

0.43

setupUi

0.40

 panel

0.38

subplot

0.38

 Мексичка

0.35

hObject

0.35

 himo

0.35

Activations Density 0.011%

references to graphical layout components in code or diagrams

panel and bottom

No Comments

No Known Activations