INDEX

Explanations

references to locations and spatial descriptions

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

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

 अभ

-0.50

出版年

-0.45

からず

-0.44

 comuniques

-0.43

ve

-0.43

crose

-0.42

维生素

-0.41

packung

-0.40

χεία

-0.39

 continúas

-0.39

POSITIVE LOGITS

 downstairs

1.69

 upstairs

1.54

 basement

1.53

 ground

1.52

ground

1.35

 Ground

1.32

 GROUND

1.27

basement

1.26

Ground

1.26

Basement

1.22

Activations Density 0.091%

references to locations and spatial descriptions

No Comments

No Known Activations