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

Попис

-0.52

windowFixed

-0.48

outState

-0.47

fromCharCode

-0.47

canestro

-0.44

 spôso

-0.44

 Biôgrafia

-0.43

 Administrativna

-0.43

posedge

-0.42

 jsPsych

-0.42

POSITIVE LOGITS

ISupport

0.56

 nahilalakip

0.53

 initComponents

0.50

IContainer

0.50

rungsseite

0.47

StoryboardSegue

0.45

 gyhoeddwyd

0.42

 AssemblyTitle

0.42

<thead>

0.42

 <<<<<<<<<<<<<<

0.41

Activations Density 0.000%

No Known Activations

This feature has no known activations.