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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 محفوظة

-0.61

ftc

-0.53

 RoutedEventArgs

-0.53

ActionResult

-0.49

 inoxydable

-0.45

 Wikiseite

-0.44

QMessageBox

-0.43

BASELINE

-0.43

 catch

-0.42

 iprot

-0.42

POSITIVE LOGITS

Initializable

1.09

rawtypes

0.97

NameInMap

0.60

 lenker

0.52

 CanadaChoose

0.50

Personensuche

0.49

 ویکی‌پدی

0.47

parsedMessage

0.47

tagHelperRunner

0.47

 BoxFit

0.46

Activations Density 0.000%

No Known Activations

This feature has no known activations.