INDEX

Explanations

references to promotions or advancements in status or rank

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

 Rond

-0.42

riente

-0.41

ListComponent

-0.39

HEIM

-0.39

λευ

-0.39

édie

-0.37

 preventivo

-0.37

 acetylcholine

-0.36

rú

-0.36

 صح

-0.36

POSITIVE LOGITS

 advancement

1.45

 upgrade

1.38

 progression

1.36

 ladder

1.35

 upward

1.35

 Progression

1.30

 upgrading

1.28

upgrade

1.26

Progression

1.26

 Advancement

1.25

Activations Density 0.253%

references to promotions or advancements in status or rank

No Comments

No Known Activations