INDEX

Explanations

mathematical symbols and notation

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

$\Phi$ or phi is called

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

 BytesLike

-0.54

expandindo

-0.54

 مشين

-0.52

Дереккөздер

-0.50

 архивлан

-0.49

marily

-0.49

HideFlags

-0.48

richTextPanel

-0.48

تقاوى

-0.47

MigrationBuilder

-0.47

POSITIVE LOGITS

 eterna

0.32

REA

0.30

bleau

0.30

 onlyOwner

0.30

าล

0.30

Periodic

0.29

stanford

0.28

stat

0.27

Jahre

0.27

 graded

0.27

Activations Density 0.400%

mathematical symbols and notation

$\Phi$ or phi is called

No Comments

No Known Activations