INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 estekak

-0.84

 snippetHide

-0.60

Diweddarwch

-0.58

ResumeLayout

-0.57

httphttps

-0.55

ंदीखरीदारी

-0.54

#+#

-0.53

 resourceCulture

-0.52

tagHelperRunner

-0.52

➟

-0.52

POSITIVE LOGITS

IDX

0.39

talet

0.38

 tulee

0.38

Folge

0.36

 depan

0.36

libft

0.35

 alent

0.35

bo

0.35

TestFixture

0.35

 caras

0.34

Activations Density 0.000%

No Known Activations

This feature has no known activations.