INDEX

Explanations

specific numerical or quantitative descriptors and conditions related to definitions, equations, or scientific measurements

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

baseline offset

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

 ujednoznacz

-0.64

WidgetItem

-0.56

 ExecuteAsync

-0.56

 MaterialApp

-0.55

XMLSchema

-0.54

astify

-0.54

يكب

-0.53

ImageIO

-0.53

 onPostExecute

-0.53

AlterField

-0.52

POSITIVE LOGITS

Spoljašnje

0.36

 embuti

0.36

 Wikispecies

0.34

Continue

0.31

tı

0.31

HCO

0.30

 VizieR

0.30

krä

0.30

 ут

0.29

læ

0.28

Activations Density 0.103%

specific numerical or quantitative descriptors and conditions related to definitions, equations, or scientific measurements

baseline offset

No Comments

No Known Activations