INDEX

Explanations

references to corrections and updates in academic articles

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

article revision status

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

########.

-0.51

SharedCtor

-0.48

Personendaten

-0.46

routeProvider

-0.43

Filmographie

-0.42

Scénario

-0.42

Tron

-0.40

WebElementEntity

-0.40

GeoNames

-0.40

OGND

-0.39

POSITIVE LOGITS

bootstrapcdn

0.57

httphttps

0.50

 honours

0.48

 resolução

0.46

 juos

0.43

 honors

0.42

期刊论文

0.42

 miniaturka

0.41

 publication

0.41

Ehrungen

0.41

Activations Density 0.006%

references to corrections and updates in academic articles

article revision status

No Comments

No Known Activations