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

RegressionTest

-0.68

expandindo

-0.65

LookAnd

-0.62

 typelib

-0.59

djangoproject

-0.59

ThroughAttribute

-0.58

SpringRunner

-0.57

findpost

-0.54

MessageTagHelper

-0.54

StoryboardSegue

-0.53

POSITIVE LOGITS

 tartalomajánló

0.49

Vorteile

0.38

 picado

0.37

⌋

0.37

 autorytatywna

0.37

Einzelnachweise

0.36

ivable

0.35

httphttps

0.34

iettivo

0.34

 мәкал

0.34

Activations Density 0.000%

No Known Activations

This feature has no known activations.