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

тинка

-0.40

 titulo

-0.39

 levier

-0.38

 itself

-0.35

 Itself

-0.34

addition

-0.34

 poivre

-0.34

Ma

-0.34

itself

-0.34

 balle

-0.34

POSITIVE LOGITS

our

1.20

Our

1.13

Our

1.13

OUR

1.07

 nuestra

0.95

our

0.94

 nossa

0.94

 nostra

0.92

 nuestras

0.90

我们的

0.89

Activations Density 0.000%

No Known Activations

This feature has no known activations.