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

FirebaseUser

-1.47

 SqlCommand

-1.22

معلومات

-0.38

 Wikiseite

-0.34

 obro

-0.32

spans

-0.31

 Hyde

-0.31

 Summe

-0.31

SqlCommand

-0.31

imageURL

-0.30

POSITIVE LOGITS

onViewCreated

4.03

getItemId

1.05

 AssemblyTitle

1.00

 onCreateView

1.00

nationality

0.95

onOptions

0.85

StoryboardSegue

0.84

onAttach

0.84

yntaxException

0.84

BufferException

0.84

Activations Density 0.000%

No Known Activations

This feature has no known activations.