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

Tikang

-0.74

 disambiguazione

-0.70

URLException

-0.64

 utafitiHapana

-0.64

CppMethod

-0.60

 يتيمه

-0.60

/*++

-0.60

mybatisplus

-0.59

IntoConstraints

-0.59

MemoryWarning

-0.56

POSITIVE LOGITS

 проєкту

3.27

 проє

1.37

 проекта

1.09

 projektu

1.09

 proiect

0.95

 Projekt

0.94

 Projektu

0.93

Projekt

0.91

 project

0.91

 PROJECT

0.90

Activations Density 0.000%

No Known Activations

This feature has no known activations.