INDEX

Explanations

terms related to medical and legal guidance

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

service providers and roles

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

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

ChildScrollView

-0.56

IntoConstraints

-0.48

ParallelGroup

-0.44



-0.41

 esternos

-0.41

OGND

-0.40

AlterField

-0.38

 причем

-0.38

GIVEREF

-0.37

 defStyleAttr

-0.37

POSITIVE LOGITS

Mys

0.49

Mys

0.45

 prescribed

0.44

 prescribes

0.42

にご

0.42

 noDo

0.41

stateProvider

0.41

dealer

0.41

󠁢

0.40

elemField

0.40

Activations Density 0.025%

terms related to medical and legal guidance

service providers and roles

No Comments

No Known Activations