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

mobileqq

-0.52

 utafitiHapana

-0.51

WebVitals

-0.48

 Counter

-0.48

 surla

-0.47

 snippetHide

-0.47

 counter

-0.45



-0.43

ucoup

-0.43

 كومونز

-0.42

POSITIVE LOGITS

disposing

1.99

MigrationBuilder

0.46

Disposed

0.44

/*

0.41

 disposing

0.40

 Dispose

0.39

Paladin

0.38

ResumeLayout

0.38

Frosty

0.37

PDIR

0.37

Activations Density 0.000%

No Known Activations

This feature has no known activations.