INDEX

Explanations

references to physical positioning and orientation in context

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

orientation

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

awtextra

-0.46

addCriterion

-0.45

windowFixed

-0.43

MigrationBuilder

-0.41

invokeLater

-0.40

الحياه

-0.40

 aéreo

-0.40

ControllerAdvice

-0.39

classnames

-0.38

stateProvider

-0.38

POSITIVE LOGITS

 flipped

0.46

 sideways

0.46

 inverted

0.44

 upside

0.44

ordion

0.43

rvore

0.43

Ƚ

0.42

pright

0.41

 rotated

0.41

__*/

0.41

Activations Density 0.061%

references to physical positioning and orientation in context

orientation

No Comments

No Known Activations