INDEX

Explanations

No Explanations Found

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

onOptions

-0.78

Tembelea

-0.75

 queſta

-0.72

#+#

-0.72

 disambiguazione

-0.71

 تكبرها

-0.71

ValueStyle

-0.70

HasForeignKey

-0.70

 صوتيه

-0.69

 سكانية

-0.69

POSITIVE LOGITS

 System

0.40

 paper

0.39

 system

0.35

 sớm

0.33

 Paper

0.32

System

0.30

 importe

0.30

UTE

0.30

 signed

0.30

 pages

0.29

Activations Density 0.000%

No Known Activations

This feature has no known activations.