INDEX

Explanations

numerical data and calculations

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

scientific notation

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

StatusOK

-0.38

CPR

-0.35

μως

-0.34

rawDesc

-0.34

CUDA

-0.33

OGND

-0.31

Aholisi

-0.30

 aikana

-0.30

Insee

-0.29

IMDG

-0.29

POSITIVE LOGITS

MemoryWarning

0.60

 Administrativna

0.55

icolon

0.50

enumii

0.50

 NACIONAL

0.49

 elefante

0.48

 OMITBAD

0.48

cipais

0.47

BagConstraints

0.46

脚注の使い方

0.46

Activations Density 0.424%

numerical data and calculations

scientific notation

No Comments

No Known Activations