INDEX

Explanations

references to achievements and milestones in various contexts

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

 Мексичка

-0.53

Portail

-0.47

 Roskov

-0.47

,:);

-0.44

assorted

-0.43

alow

-0.42

Життєпис

-0.42

})()

-0.41

>",

-0.41

/>";

-0.41

POSITIVE LOGITS

 first

0.49

AndEndTag

0.48

ServletException

0.47

 fourth

0.45

 final

0.45

UnusedPrivate

0.45

 StatelessWidget

0.44

 ever

0.44

 fifth

0.44

tangentMode

0.44

Activations Density 0.177%

references to achievements and milestones in various contexts

No Comments

No Known Activations