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

[]:

-0.38

TableModel

-0.38

 Schen

-0.37

 distant

-0.36

 hial

-0.35

 Coul

-0.35

 Island

-0.34

 same

-0.34

 Same

-0.34

Zit

-0.34

POSITIVE LOGITS

ButterKnife

5.25

 butterknife

0.83

 HttpNotFound

0.75

BufferException

0.70

 فريبيس

0.69

IntoConstraints

0.69

 gynhyrchwyd

0.68

setupUi

0.66

BindView

0.64

GOTREF

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.