INDEX

Explanations

mathematical symbols and structures in a context involving equations or expressions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.12.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

ness

-0.88

ers

-0.79

er

-0.75

—

-0.70

an

-0.70

ja

-0.68

Del

-0.68

ating

-0.67

ern

-0.67

{-\

-0.67

POSITIVE LOGITS

]")]

1.34

"}

1.29

+#+#

1.27

 }}$}

1.26

"]}

1.23

']}

1.20

'}

1.20

")}

1.14

"}

1.03

";}

1.00

Activations Density 0.150%

mathematical symbols and structures in a context involving equations or expressions

No Comments

No Known Activations