INDEX

Explanations

patterns involving numerical values or sequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 39-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_39/width_16k/average_l0_131

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.39.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

 terciopelo

-0.41

Xem

-0.40

Voci

-0.39

rowse

-0.39

 tebal

-0.39

[*]

-0.38

Estas

-0.38

AddTagHelper

-0.38

OT

-0.37

 ursprünglichen

-0.36

POSITIVE LOGITS

 Италијани

0.48

};*/

0.46

'{@

0.46

 autorytatywna

0.45

NameInMap

0.44

 androidx

0.44

*/;

0.43

/>";

0.42

 aapt

0.42

 kangoo

0.41

Activations Density 0.149%

patterns involving numerical values or sequences

No Comments

No Known Activations