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

CppMethod

-0.49

expandindo

-0.49

 JpaRepository

-0.47

RegressionTest

-0.42

 chuva

-0.39

ImageContext

-0.38

 Backman

-0.37

 inilah

-0.37

esModule

-0.37

 emisión

-0.37

POSITIVE LOGITS

хьтан

0.56

 transfieras

0.53

 nakalista

0.50

EndContext

0.47

 spacious

0.45

 متعلقه

0.45

BeginInit

0.44

Sqft

0.43

openzeppelin

0.42

Activations Density 0.000%

No Known Activations

This feature has no known activations.