INDEX

Explanations

specific characters or symbols in a sequence related to 'J'

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_33/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.33.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

出版年

-0.43

expandindo

-0.40

 دیکھیے

-0.40

 يتيمه

-0.39

 lote

-0.39

 sede

-0.38

Overrides

-0.38

 rangs

-0.37

 Breaking

-0.36

AUF

-0.36

POSITIVE LOGITS

帖最后由

0.58

}$

0.48

ंदीखरीदारी

0.42

 verschillen

0.42

argon

0.42

󠁢

0.42

╝

0.42

)))))

0.42

 BoxDecoration

0.42

 gezondheid

0.40

Activations Density 0.272%

specific characters or symbols in a sequence related to 'J'

No Comments

No Known Activations