INDEX

Explanations

references to numerical data or statistics

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

DR values

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

+#+#

-0.52

="@+

-0.49

 ModelExpression

-0.47

 tartalomajánló

-0.45

SequentialGroup

-0.43

 arşivlendi

-0.42

 AssemblyCulture

-0.42

WriteBarrier

-0.41

 Référence

-0.41

 StatelessWidget

-0.41

POSITIVE LOGITS

colgante

0.49

 kostenlosen

0.47

convite

0.45

idać

0.44

 dokun

0.43

hilangan

0.42

caneca

0.42

 Dimensiones

0.41

disfraz

0.41

 recoger

0.41

Activations Density 0.253%

references to numerical data or statistics

DR values

No Comments

No Known Activations