INDEX

Explanations

numerical data related to study populations and outcomes

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

numbers and percentages

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

aarrggbb

-0.50

 Italijanski

-0.49

extAlignment

-0.45

énario

-0.45

TypedDataSet

-0.45

nodoc

-0.44

addCriterion

-0.42

Sob

-0.42

plr

-0.42

 Италијани

-0.41

POSITIVE LOGITS

fører

0.49

 الحره

0.49

 للاسماء

0.46

UserController

0.41

 autorytatywna

0.41

Empereur

0.41

 indietro

0.41

garek

0.39

 tilbage

0.39

føring

0.39

Activations Density 0.064%

numerical data related to study populations and outcomes

numbers and percentages

No Comments

No Known Activations