INDEX

Explanations

technical terms and concepts related to statistical modeling and programming

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

"proportional hazards"

np_acts-logits-general · gemini-2.0-flash

Cox proportional hazards

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

ValueStyle

-0.58

MessageWindow

-0.57

ſſung

-0.57

Bakgrunnsstoff

-0.57

Personensuche

-0.57

aarrggbb

-0.56

ロウィン

-0.56

 queſto

-0.53

createClass

-0.53

DockStyle

-0.53

POSITIVE LOGITS

InSection

0.30

attaque

0.30

 respekt

0.29

awtextra

0.29

친

0.29

 intéressant

0.29

ţiei

0.28

flich

0.28

PORARY

0.28

SHI

0.28

Activations Density 0.540%

technical terms and concepts related to statistical modeling and programming

"proportional hazards"

Cox proportional hazards

No Comments

No Known Activations