INDEX

Explanations

numerical data or metrics associated with scientific or quantitative analysis

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

number comparisons

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

ped

-0.45

 government

-0.34

…

-0.34

 configureStore

-0.34

getValues

-0.30

 pand

-0.29

…”

-0.29

 roule

-0.29

Employees

-0.29

 COAST

-0.29

POSITIVE LOGITS

AndEndTag

0.71

 iNdEx

0.62

 متعلقه

0.59

 oprot

0.56

RegressionTest

0.55

 ujednoznacz

0.55

ExecuteReader

0.55

 defaultstate

0.52

 AttributeSet

0.52

styleType

0.50

Activations Density 0.263%

numerical data or metrics associated with scientific or quantitative analysis

number comparisons

No Comments

No Known Activations