INDEX

Explanations

programming constructs and data types used in software development

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

action event or object wrapper

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

 nemlig

-0.52

SequentialGroup

-0.49

 totiž

-0.47

sätzlich

-0.43

PARATUS

-0.41

 bowiem

-0.41

 desmotivaciones

-0.40

 isInitialized

-0.40

 zoude

-0.39

 infatti

-0.39

POSITIVE LOGITS

enumi

0.59

="@+

0.59

 AttributeSet

0.57

BuildContext

0.54

 astore

0.54

ruptedException

0.51

WebMethod

0.49

 RoutedEventArgs

0.48

tsin

0.47

MigrationBuilder

0.47

Activations Density 0.039%

programming constructs and data types used in software development

action event or object wrapper

No Comments

No Known Activations