INDEX

Explanations

references to specific concepts or ideas, particularly those represented by the word "it."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

/');

-0.52

Moc

-0.51

 strani

-0.51

cura

-0.49

 الذين

-0.49

 apie

-0.49

er

-0.48

edas

-0.47

 którzy

-0.47

/");

-0.47

POSITIVE LOGITS

OGND

0.89

 estekak

0.82

 Biôgrafia

0.78

RTEE

0.75

 للاسماء

0.73

seigneur

0.70

 ujednoznacz

0.70

WebVitals

0.70

צלחה

0.70

Gön

0.67

Activations Density 0.146%

references to specific concepts or ideas, particularly those represented by the word "it."

No Comments

No Known Activations