INDEX

Explanations

names and geographic entities associated with organizations or individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 BoxFit

-1.02

 мәкал

-0.94

 виправивши

-0.89

Personendaten

-0.89

ViewFeatures

-0.86

Autoritní

-0.83

 Италијани

-0.82

 autorytatywna

-0.82

 estekak

-0.81

 varandra

-0.81

POSITIVE LOGITS

 Swedish

0.83

 schwed

0.82

Swedish

0.79

 Sweden

0.74

 swedish

0.73

Sweden

0.71

Fredrik

0.69

 Lund

0.68

瑞典

0.67

 Gothenburg

0.67

Activations Density 0.643%

names and geographic entities associated with organizations or individuals

No Comments

No Known Activations