INDEX

Explanations

terms related to medical conditions and treatments

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

existing enclosingScope

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

 مشين

-0.65



-0.60

ViewImports

-0.56

 oprot

-0.56

SharedDtor

-0.52

ConstraintMaker

-0.50

OGND

-0.47

\{\\

-0.47

kháu

-0.45

ðsíða

-0.45

POSITIVE LOGITS

RectangleBorder

0.40

MigrationBuilder

0.38

Max

0.37

krieg

0.37

ẫ

0.36

saraba

0.35

Interpolator

0.35

IContainer

0.35

lek

0.34

 sense

0.33

Activations Density 0.135%

terms related to medical conditions and treatments

existing enclosingScope

No Comments

No Known Activations