INDEX

Explanations

geographical locations, particularly names of places and regions

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

 Arque

-0.68

 Jurí

-0.59

 item

-0.56

 EClass

-0.55

 Piscina

-0.54

 Audiodateien

-0.52

 côn

-0.52

 Shrimp

-0.52

Identyfik

-0.51

 Dred

-0.51

POSITIVE LOGITS

 Ontario

1.52

 Quebec

1.40

Ontario

1.38

 Manitoba

1.35

 Yorkshire

1.35

 Scotland

1.33

 Alberta

1.32

 Scottish

1.29

 Colorado

1.28

 Saskatchewan

1.27

Activations Density 2.612%

geographical locations, particularly names of places and regions

No Comments

No Known Activations