INDEX

Explanations

references to geographical locations and regions in Australia

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

Prepositions and symbols followed by a new word

np_acts-logits-general · gemini-2.0-flash

Supreme Court

np_acts-logits-general · gemini-2.5-flash-lite

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

gorilla

-0.52

ográficas

-0.44

GetAxis

-0.43

 Champlain

-0.43

NewGuid

-0.43

UIAlert

-0.39

Xaml

-0.39

SourceChecksum

-0.38

 зовні

-0.38

 Thái

-0.38

POSITIVE LOGITS

OGND

0.85

 يتيمه

0.77

})));

0.68

SharedCtor

0.68

PerPage

0.68

 Adelaide

0.65

Hentet

0.65

 تضيفلها

0.64

 Saha

0.63

})$}

0.62

Activations Density 0.052%

references to geographical locations and regions in Australia

Prepositions and symbols followed by a new word

Supreme Court

No Comments

No Known Activations