INDEX

Explanations

occurrences of the word "for."

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

purpose or duration

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

KommentareTeilen

-0.49

WebElementEntity

-0.44

 okuyayım

-0.40

toy

-0.39

*/),

-0.39

\]

-0.38

}>;

-0.37

 Burton

-0.37

toy

-0.37

AlterField

-0.37

POSITIVE LOGITS

awtextra

0.59

 loob

0.48

 ویکی‌پدی

0.47

brainly

0.47

 سكانية

0.45

因为它

0.45

GEBURTSDATUM

0.43

 kerana

0.43

 fince

0.42

ajuku

0.42

Activations Density 0.088%

occurrences of the word "for."

purpose or duration

No Comments

No Known Activations