INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

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

 surla

-0.50



-0.45

ValueStyle

-0.43

localctx

-0.42

FailureListener

-0.40

 ujarnya

-0.38

clerview

-0.35

Dichloropropane

-0.35

 CreateTagHelper

-0.35

XtraEditors

-0.34

POSITIVE LOGITS

Personensuche

0.60

 ویکی‌پدیا

0.58

GEBURTS

0.53

 beginnetje

0.50

 للاسماء

0.48

 initComponents

0.48

 autorytatywna

0.46

 resourceCulture

0.46

 toxins

0.44

Dtor

0.44

Activations Density 0.000%

No Known Activations

This feature has no known activations.