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

IBOutlet

-0.53

Bia

-0.51

shine

-0.50

Heil

-0.49

ligh

-0.49

synthesize

-0.49

 winds

-0.49

Bia

-0.48

olation

-0.48

 фор

-0.48

POSITIVE LOGITS

 DriverManager

2.84

DriverManager

1.57

 تضيفلها

0.70

Personendaten

0.48

mybatisplus

0.48

 vandens

0.45

 näin

0.43

 TestBed

0.42

 pertanyaan

0.41

wahati

0.40

Activations Density 0.000%

No Known Activations

This feature has no known activations.