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

med

-0.39

 Barlow

-0.39

 lived

-0.38

 Adkins

-0.38

 Beasley

-0.38

 Tapia

-0.37

 gazed

-0.36

head

-0.36

 Gallo

-0.35

chauen

-0.35

POSITIVE LOGITS

tagHelperRunner

0.74

 CreateTagHelper

0.71

 nakalista

0.69

StackSize

0.67

 noDo

0.63

0.62

 nahilalakip

0.58

ніципа

0.57

 estekak

0.57

ьаж

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.