INDEX

Explanations

phrases and recommendations encouraging action or awareness in users

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 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

 autorytatywna

-0.67

 ModelExpression

-0.60

 للمعارف

-0.54

 ReactDOM

-0.53

createStatement

-0.52

cæ

-0.51

RetentionPolicy

-0.50

 MATER

-0.50

 GIPHY

-0.49

localctx

-0.49

POSITIVE LOGITS

 check

0.49

 consider

0.48

consider

0.44

check

0.42

 remember

0.42

 consult

0.42

remember

0.41

 memperhatikan

0.41

 note

0.40

 Consulte

0.39

Activations Density 0.263%

phrases and recommendations encouraging action or awareness in users

No Comments

No Known Activations