INDEX

Explanations

technical terms and code-related elements

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

astéro

-0.55

 defaultstate

-0.52

دانشنامهٔ

-0.50

AndroidJUnit

-0.49

Vidite

-0.49

EDEFAULT

-0.45

 StatelessWidget

-0.44

styleType

-0.44

出版年

-0.43

 opérés

-0.43

POSITIVE LOGITS

Brainz

0.42

#+#

0.40

PickerController

0.38

Predecesor

0.37

sschutz

0.37

CustomAttributes

0.36

Sponsored

0.34

Cyfarwyddwr

0.34

 tych

0.34

***

0.33

Activations Density 1.456%

technical terms and code-related elements

No Comments

No Known Activations