INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

Cyfeiriadau

-0.74

GEBURTSDATUM

-0.71

jupiter

-0.71

Посилання

-0.64

aternion

-0.63

Rüyada

-0.63

addContainerGap

-0.60

Faust

-0.60

 Pessoa

-0.59

Бележки

-0.58

POSITIVE LOGITS

 queſta

0.77

uestamente

0.76

 Scro

0.70

erschiedene

0.70

Squ

0.69

 pilgri

0.68

GHIJKLM

0.67

éphane

0.67

Swe

0.66

 Thri

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.