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

Sah

-0.43

fat

-0.36

Yon

-0.36

ioutil

-0.36

SSC

-0.35

Koh

-0.35

.•

-0.35

ɵ

-0.35

Sis

-0.35

Ala

-0.35

POSITIVE LOGITS

RequestMapping

2.00

WebServlet

0.90

GetMapping

0.81

 propOrder

0.68

antMatchers

0.65

RectangleBorder

0.63

ArgumentParser

0.60

 AssemblyTitle

0.59

 miniaturka

0.59

XmlAccessorType

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.