INDEX

Explanations

specific formatting or notation related to programming or structured data presentations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

oa̍t

-0.49

"]();

-0.44

 Mannes

-0.43

 hochzeit

-0.43

 noqa

-0.42

Ero

-0.42

 protoimpl

-0.40

pieler

-0.40

vux

-0.40

 perfection

-0.40

POSITIVE LOGITS

 للمعارف

0.42

 tegens

0.33

 GenerationType

0.33

 JpaRepository

0.32

 MessageBoxIcon

0.32

ContentLoaded

0.32

 referenties

0.32

 resourceCulture

0.32

uppa

0.31

 XCTest

0.31

Activations Density 0.001%

specific formatting or notation related to programming or structured data presentations

No Comments

No Known Activations