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

didSet

-0.45

flat

-0.42

<?

-0.42

帖最后由

-0.41

__.__

-0.41

UnsafeEnabled

-0.41

 Mera

-0.40

 zucchini

-0.40

 ioutil

-0.39

sizeCache

-0.39

POSITIVE LOGITS

RequestMapping

2.13

GetMapping

0.77

ViewFeatures

0.60

PostMapping

0.57

 صوتيه

0.50

 Numerade

0.48

 keres

0.46

 TEACHING

0.45

 autorytatywna

0.45

verwijspagina

0.45

Activations Density 0.000%

No Known Activations

This feature has no known activations.