INDEX

Explanations

instances of political figures and terms associated with legal or formal contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

SpringRunner

-0.62

ฉัน

-0.60

ChildScrollView

-0.58

lagt

-0.58

komo

-0.57

 gesteld

-0.57

kasse

-0.57

 trá

-0.55

 Schulte

-0.55

 годов

-0.54

POSITIVE LOGITS

migrationBuilder

0.86

LLocation

0.76

 CommonModule

0.73

ousness

0.73

 autorytatywna

0.73

ية

0.72

hood

0.68

ous

0.67

theore

0.66

AccessFile

0.65

Activations Density 0.598%

instances of political figures and terms associated with legal or formal contexts

No Comments

No Known Activations