INDEX

Explanations

references to quantitative data and measurement in studies

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

SharedDtor

-0.40

tgär

-0.40

 nghiệp

-0.40

ross

-0.36

 infecciones

-0.35

 rodent

-0.35

 resulting

-0.35

ArrowToggle

-0.34

urnia

-0.34

出版年

-0.33

POSITIVE LOGITS

 because

0.81

 omdat

0.70

 porque

0.69

because

0.69

 потому

0.68

 Because

0.63

Because

0.63

あえて

0.63

 perchè

0.62

 karena

0.62

Activations Density 0.800%

references to quantitative data and measurement in studies

No Comments

No Known Activations