INDEX

Explanations

many words that are prepositions, articles, and words indicating location

oai_token-act-pair · gemini-2.0-flash

geographical regions

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

новништво

-0.52

atche

-0.47

 gratuites

-0.46

 Tasmania

-0.43

 Israel

-0.43

 France

-0.42

?";

-0.42

 Freiw

-0.42

 Nigeria

-0.41

France

-0.41

POSITIVE LOGITS

 Asian

1.60

Asian

1.43

 European

1.27

 Asians

1.25

 asian

1.25

 Southeast

1.24

 Eastern

1.16

 African

1.13

Southeast

1.13

 Latin

1.12

Activations Density 1.331%

many words that are prepositions, articles, and words indicating location

geographical regions

No Comments

No Known Activations