INDEX

Explanations

mathematical or technical notation and symbols

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 mapStateToProps

-0.45

bul

-0.44

 Fork

-0.43

extAlignment

-0.42

Hentet

-0.42

 satisfaction

-0.41

 flat

-0.41

dcmtk

-0.41

 simultaneous

-0.41

 localObject

-0.41

POSITIVE LOGITS

ConstraintMaker

0.65

Personensuche

0.56

SharedCtor

0.49

//

0.45

IComponent

0.44

Carriera

0.43

/**

0.42

LookAnd

0.42

 चीज़ों

0.40

Activations Density 0.046%

mathematical or technical notation and symbols

No Comments

No Known Activations