INDEX

Explanations

references to specific geographical locations and individuals associated with India

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

Bronnen

-0.65

 Matsu

-0.55

 Japanese

-0.52

dó

-0.51

Mih

-0.50

Yok

-0.48

 drift

-0.48

Oth

-0.47

 gant

-0.47

Mey

-0.47

POSITIVE LOGITS

 India

1.42

India

1.41

 Indian

1.36

 Mumbai

1.34

 Bangalore

1.32

 Delhi

1.30

Mumbai

1.29

 DELHI

1.25

Indian

1.25

 INDIA

1.25

Activations Density 2.776%

references to specific geographical locations and individuals associated with India

No Comments

No Known Activations